code
stringlengths 82
54.1k
| code_codestyle
int64 0
699
| style_context
stringlengths 111
35.6k
| style_context_codestyle
int64 0
699
| label
int64 0
1
|
|---|---|---|---|---|
import json
import os
import unittest
from typing import Tuple
from transformers import WavaVecaPhonemeCTCTokenizer
from transformers.models.wavaveca.tokenization_wavaveca import VOCAB_FILES_NAMES
from transformers.models.wavaveca_phoneme.tokenization_wavaveca_phoneme import WavaVecaPhonemeCTCTokenizerOutput
from transformers.testing_utils import require_phonemizer
from ...test_tokenization_common import TokenizerTesterMixin
@require_phonemizer
class _a ( UpperCAmelCase__ , unittest.TestCase ):
"""simple docstring"""
A_ = WavaVecaPhonemeCTCTokenizer
A_ = False
def _UpperCAmelCase ( self ) -> Optional[int]:
super().setUp()
UpperCamelCase_ = (
'<s> <pad> </s> <unk> n s t ə l a i k d m ɛ ɾ e ɪ p o ɐ z ð f j v b ɹ ʁ ʊ iː r w ʌ u ɡ æ aɪ ʃ h ɔ ɑː '
'ŋ ɚ eɪ β uː y ɑ̃ oʊ ᵻ eː θ aʊ ts oː ɔ̃ ɣ ɜ ɑ dʒ əl x ɜː ç ʒ tʃ ɔː ɑːɹ ɛ̃ ʎ ɔːɹ ʋ aː ɕ œ ø oːɹ ɲ yː '
'ʔ iə i5 s. tɕ ?? nʲ ɛː œ̃ ɭ ɔø ʑ tʲ ɨ ɛɹ ts. rʲ ɪɹ ɭʲ i.5 ɔɪ q sʲ u5 ʊɹ iɜ a5 iɛ5 øː ʕ ja əɜ th ɑ5 '
'oɪ dʲ ə5 tɕh ts.h mʲ ɯ dʑ vʲ e̞ tʃʲ ei5 o5 onɡ5 ɑu5 iɑ5 ai5 aɪɚ kh ə1 ʐ i2 ʉ ħ t[ aɪə ʲ ju ə2 u2 oɜ '
'pː iɛɜ ou5 y5 uɜ tː uo5 d[ uoɜ tsh ɑɜ ɵ i̪5 uei5 ɟ aɜ ɑɨ i.ɜ eʊ o2 ɐ̃ ä pʲ kʲ n̩ ɒ ph ɑu2 uɨ əɪ ɫ ɬ '
'yɜ bʲ ɑ2 s̪ aiɜ χ ɐ̃ʊ̃ 1 ə4 yæɜ a2 ɨː t̪ iouɜ ũ onɡɜ aɨ iɛ2 ɔɨ ɑuɜ o̞ ei2 iou2 c kː y2 ɖ oe dˤ yɛɜ '
'əʊ S ɡʲ onɡ2 u" eiɜ ʈ ɯᵝ iou5 dZ r̝̊ i.2 tS s^ ʝ yə5 iɑɜ uə5 pf ɨu iɑ2 ou2 ər2 fʲ ai2 r̝ uəɜ ɳ əɨ '
'ua5 uɪ ɽ bː yu5 uo2 yɛ5 l̩ ɻ ərɜ ʂ i̪2 ouɜ uaɜ a. a.ː yæ5 dː r̩ ee ɪu ər5 i̪ ɜ æi u: i.ː t^ o1 ɪ^ '
'ai ueiɜ æː ɛɪ eə i. ɴ ie ua2 ɑ1 o4 tʃː o: ɑ: u1 N i̪1 au yæ2 u. qː yəɜ y: kʰ tʃʰ iʊ sx õ uo tʰ '
'uai5 bʰ u.ː uə2 ʊə d^ s̪ː yiɜ dʰ r. oe: i1 ɟː yu2 nʲʲ i̪4 uei2 tsʲ ɸ ĩ ɑ4 t̪ː eɑ u4 e: tsː ʈʰ ɡʰ '
'ɯɯ dʒʲ ʂʲ X ɵː uaiɜ tɕʲ ã t^ː ẽː yɛ2 cː i.1 ɛʊ dˤdˤ dʒː i4 ɡː yi ɕʲ ɟʰ pʰ dʑʲ yuɜ ua1 ua4 æiː ɐɐ '
'ui iou1 ʊː a1 iou4 cʰ iɛ1 yə2 ɖʰ ẽ ʒʲ ää ər4 iːː ɪː iɑ1 ər1 œː øi ɪuː cʰcʰ əː1 iː1 ũ kʰː o̞o̞ xʲ '
'ou1 iɛ4 e̞e̞ y1 dzː dʲʲ dʰː ɯᵝɯᵝ lː uo1 i.4 i: yɛ5ʲ a4'
).split(' ' )
UpperCamelCase_ = dict(zip(_UpperCAmelCase , range(len(_UpperCAmelCase ) ) ) )
UpperCamelCase_ = {'pad_token': '<pad>', 'unk_token': '<unk>', 'bos_token': '<s>', 'eos_token': '</s>'}
UpperCamelCase_ = os.path.join(self.tmpdirname , VOCAB_FILES_NAMES['vocab_file'] )
with open(self.vocab_file , 'w' , encoding='utf-8' ) as fp:
fp.write(json.dumps(_UpperCAmelCase ) + '\n' )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase=False , _UpperCAmelCase=20 , _UpperCAmelCase=5 ) -> Tuple[str, list]:
UpperCamelCase_ = [(i, tokenizer.decode([i] , clean_up_tokenization_spaces=_UpperCAmelCase )) for i in range(len(_UpperCAmelCase ) )]
UpperCamelCase_ = list(filter(lambda _UpperCAmelCase : [t[0]] == tokenizer.encode(t[1] , do_phonemize=_UpperCAmelCase ) , _UpperCAmelCase ) )
if max_length is not None and len(_UpperCAmelCase ) > max_length:
UpperCamelCase_ = toks[:max_length]
if min_length is not None and len(_UpperCAmelCase ) < min_length and len(_UpperCAmelCase ) > 0:
while len(_UpperCAmelCase ) < min_length:
UpperCamelCase_ = toks + toks
# toks_str = [t[1] for t in toks]
UpperCamelCase_ = [t[0] for t in toks]
# Ensure consistency
UpperCamelCase_ = tokenizer.decode(_UpperCAmelCase , clean_up_tokenization_spaces=_UpperCAmelCase )
if " " not in output_txt and len(_UpperCAmelCase ) > 1:
UpperCamelCase_ = (
tokenizer.decode([toks_ids[0]] , clean_up_tokenization_spaces=_UpperCAmelCase )
+ ' '
+ tokenizer.decode(toks_ids[1:] , clean_up_tokenization_spaces=_UpperCAmelCase )
)
if with_prefix_space:
UpperCamelCase_ = ' ' + output_txt
UpperCamelCase_ = tokenizer.encode(_UpperCAmelCase , add_special_tokens=_UpperCAmelCase )
return output_txt, output_ids
def _UpperCAmelCase ( self , **_UpperCAmelCase ) -> Optional[int]:
kwargs.update(self.special_tokens_map )
return WavaVecaPhonemeCTCTokenizer.from_pretrained(self.tmpdirname , **_UpperCAmelCase )
def _UpperCAmelCase ( self ) -> Dict:
UpperCamelCase_ = self.tokenizer_class.from_pretrained('facebook/wav2vec2-lv-60-espeak-cv-ft' )
# check adding a single token
tokenizer.add_tokens('xxx' )
UpperCamelCase_ = tokenizer('m xxx ɪ' , do_phonemize=_UpperCAmelCase ).input_ids
self.assertEqual(_UpperCAmelCase , [13, 392, 17] ) # xxx should be last token
tokenizer.add_tokens(['aaa', 'bbb', 'ccc'] )
UpperCamelCase_ = tokenizer('m aaa ɪ ccc' , do_phonemize=_UpperCAmelCase ).input_ids
self.assertEqual(_UpperCAmelCase , [13, 393, 17, 395] ) # aaa and ccc should be after xxx and 2 after aaa
UpperCamelCase_ = tokenizer('maɪ c' , do_phonemize=_UpperCAmelCase ).input_ids
self.assertEqual(_UpperCAmelCase , [3, 200] ) # mai should be <unk> (=3)
def _UpperCAmelCase ( self ) -> List[Any]:
UpperCamelCase_ = self.tokenizer_class.from_pretrained('facebook/wav2vec2-lv-60-espeak-cv-ft' )
UpperCamelCase_ = 'Hello how are you'
UpperCamelCase_ = tokenizer.phonemize(_UpperCAmelCase , phonemizer_lang='en-us' )
self.assertEqual(_UpperCAmelCase , 'h ə l oʊ h aʊ ɑːɹ j uː' )
def _UpperCAmelCase ( self ) -> Tuple:
UpperCamelCase_ = self.tokenizer_class.from_pretrained('facebook/wav2vec2-lv-60-espeak-cv-ft' )
UpperCamelCase_ = 'Hello how are you'
UpperCamelCase_ = tokenizer.phonemize(_UpperCAmelCase , phonemizer_lang='en-us' )
self.assertEqual(tokenizer(_UpperCAmelCase ).input_ids , tokenizer(_UpperCAmelCase , do_phonemize=_UpperCAmelCase ).input_ids )
def _UpperCAmelCase ( self ) -> List[Any]:
UpperCamelCase_ = self.tokenizer_class.from_pretrained('facebook/wav2vec2-lv-60-espeak-cv-ft' )
UpperCamelCase_ = 'Hello how are you'
UpperCamelCase_ = tokenizer.phonemize(_UpperCAmelCase , phonemizer_lang='en-us' )
UpperCamelCase_ = tokenizer.decode(tokenizer(_UpperCAmelCase ).input_ids )
self.assertEqual(_UpperCAmelCase , _UpperCAmelCase )
def _UpperCAmelCase ( self ) -> str:
UpperCamelCase_ = self.tokenizer_class.from_pretrained('facebook/wav2vec2-lv-60-espeak-cv-ft' )
UpperCamelCase_ = [
[11, 5, 15, tokenizer.pad_token_id, 15, 8, 98],
[24, 22, 5, 24, 22, 5, 77],
]
UpperCamelCase_ = tokenizer.decode(sample_ids[0] )
UpperCamelCase_ = tokenizer.batch_decode(_UpperCAmelCase )
self.assertEqual(_UpperCAmelCase , batch_tokens[0] )
self.assertEqual(_UpperCAmelCase , ['k s ɾ ɾ l ɭʲ', 'j ð s j ð s oːɹ'] )
def _UpperCAmelCase ( self ) -> int:
UpperCamelCase_ = self.tokenizer_class.from_pretrained(
'facebook/wav2vec2-lv-60-espeak-cv-ft' , word_delimiter_token='|' )
tokenizer.add_tokens('|' )
UpperCamelCase_ = 'Hello how are you'
UpperCamelCase_ = tokenizer.phonemize(_UpperCAmelCase , phonemizer_lang='en-us' )
self.assertEqual(_UpperCAmelCase , 'h ə l oʊ | h aʊ | ɑːɹ | j uː |' )
def _UpperCAmelCase ( self ) -> Any:
UpperCamelCase_ = self.tokenizer_class.from_pretrained(
'facebook/wav2vec2-lv-60-espeak-cv-ft' , word_delimiter_token='|' )
tokenizer.add_tokens('|' )
UpperCamelCase_ = 'Hello how are you'
UpperCamelCase_ = tokenizer.phonemize(_UpperCAmelCase , phonemizer_lang='en-us' )
self.assertEqual(tokenizer(_UpperCAmelCase ).input_ids , tokenizer(_UpperCAmelCase , do_phonemize=_UpperCAmelCase ).input_ids )
def _UpperCAmelCase ( self ) -> Tuple:
UpperCamelCase_ = self.tokenizer_class.from_pretrained(
'facebook/wav2vec2-lv-60-espeak-cv-ft' , word_delimiter_token='|' )
tokenizer.add_tokens('|' )
# fmt: off
UpperCamelCase_ = [
[11, 5, 15, tokenizer.pad_token_id, tokenizer.word_delimiter_token_id, 15, 8, tokenizer.word_delimiter_token_id, 98],
[tokenizer.word_delimiter_token_id, 24, 22, tokenizer.word_delimiter_token_id, 5, 24, 22, 5, 77],
]
# fmt: on
# decode with word_del_token filter
UpperCamelCase_ = tokenizer.decode(sample_ids[0] )
UpperCamelCase_ = tokenizer.batch_decode(_UpperCAmelCase )
self.assertEqual(_UpperCAmelCase , batch_tokens[0] )
self.assertEqual(_UpperCAmelCase , ['k s ɾ ɾ l ɭʲ', 'j ð s j ð s oːɹ'] )
# decode with no word_del_token filter
UpperCamelCase_ = tokenizer.decode(sample_ids[0] , filter_word_delimiter_token=_UpperCAmelCase )
UpperCamelCase_ = tokenizer.batch_decode(_UpperCAmelCase , filter_word_delimiter_token=_UpperCAmelCase )
self.assertEqual(_UpperCAmelCase , batch_tokens[0] )
self.assertEqual(_UpperCAmelCase , ['k s ɾ | ɾ l | ɭʲ', '| j ð | s j ð s oːɹ'] )
def _UpperCAmelCase ( self ) -> List[Any]:
UpperCamelCase_ = self.tokenizer_class.from_pretrained(
'facebook/wav2vec2-lv-60-espeak-cv-ft' , word_delimiter_token='|' )
tokenizer.add_tokens('|' )
UpperCamelCase_ = 'Hello how are you'
UpperCamelCase_ = tokenizer.phonemize(_UpperCAmelCase , phonemizer_lang='en-us' )
UpperCamelCase_ = tokenizer.decode(tokenizer(_UpperCAmelCase ).input_ids , filter_word_delimiter_token=_UpperCAmelCase )
self.assertEqual(_UpperCAmelCase , _UpperCAmelCase )
def _UpperCAmelCase ( self ) -> Optional[int]:
UpperCamelCase_ = self.tokenizer_class.from_pretrained(
'facebook/wav2vec2-lv-60-espeak-cv-ft' , word_delimiter_token='|' )
tokenizer.add_tokens('|' )
UpperCamelCase_ = 'Hello how are you'
UpperCamelCase_ = tokenizer.phonemize(_UpperCAmelCase , phonemizer_lang='en-us' )
UpperCamelCase_ = tokenizer.decode(tokenizer(_UpperCAmelCase ).input_ids , filter_word_delimiter_token=_UpperCAmelCase )
self.assertEqual(' '.join([p.strip() for p in phonemes.split(' |' )] ).strip() , _UpperCAmelCase )
def _UpperCAmelCase ( self ) -> Any:
UpperCamelCase_ = self.tokenizer_class.from_pretrained(
'facebook/wav2vec2-lv-60-espeak-cv-ft' , word_delimiter_token=_UpperCAmelCase )
UpperCamelCase_ = 'Hello how are you'
UpperCamelCase_ = tokenizer(_UpperCAmelCase , phonemizer_lang='en-us' ).input_ids
UpperCamelCase_ = tokenizer(_UpperCAmelCase , phonemizer_lang='fr-fr' ).input_ids
self.assertNotEqual(_UpperCAmelCase , _UpperCAmelCase )
UpperCamelCase_ = tokenizer.decode(_UpperCAmelCase )
UpperCamelCase_ = tokenizer.decode(_UpperCAmelCase )
self.assertEqual(_UpperCAmelCase , 'h ə l oʊ h aʊ ɑːɹ j uː' )
self.assertEqual(_UpperCAmelCase , 'ɛ l o h aʊ a ʁ j u' )
def _UpperCAmelCase ( self ) -> Union[str, Any]:
UpperCamelCase_ = self.tokenizer_class.from_pretrained('facebook/wav2vec2-lv-60-espeak-cv-ft' )
UpperCamelCase_ = 'Hello how Are you'
UpperCamelCase_ = 'hello how are you'
UpperCamelCase_ = tokenizer(_UpperCAmelCase ).input_ids
UpperCamelCase_ = tokenizer(_UpperCAmelCase ).input_ids
self.assertEqual(_UpperCAmelCase , _UpperCAmelCase )
def _UpperCAmelCase ( self ) -> Union[str, Any]:
UpperCamelCase_ = self.tokenizer_class.from_pretrained('facebook/wav2vec2-lv-60-espeak-cv-ft' )
tokenizer.add_tokens(['!', '?'] )
tokenizer.add_special_tokens({'cls_token': '$$$'} )
# fmt: off
UpperCamelCase_ = [
[11, 5, 15, tokenizer.pad_token_id, 15, 8, 98, 392, 392, 393, 392, 392, 393, 394, 394],
[24, 22, 5, 24, 22, 5, 77, tokenizer.pad_token_id, 394, 394],
]
# fmt: on
UpperCamelCase_ = tokenizer.batch_decode(_UpperCAmelCase )
self.assertEqual(_UpperCAmelCase , ['k s ɾ ɾ l ɭʲ!?!? $$$', 'j ð s j ð s oːɹ $$$'] )
@staticmethod
def _UpperCAmelCase ( _UpperCAmelCase , _UpperCAmelCase ) -> Union[str, Any]:
UpperCamelCase_ = [d[key] for d in offsets]
return retrieved_list
def _UpperCAmelCase ( self ) -> Union[str, Any]:
UpperCamelCase_ = self.get_tokenizer(word_delimiter_token='|' )
tokenizer.add_tokens('|' )
# fmt: off
# ksssɾɾ|ɾɾ<pad>ɾɾ|<pad>ɾlll|ɭʲ -> k s ɾ ɾ | ɾ l | ɭʲ"
UpperCamelCase_ = [11, 5, 5, 5, 15, 15, tokenizer.pad_token_id, 15, 15, tokenizer.word_delimiter_token_id, tokenizer.pad_token_id, 15, 8, 8, 8, tokenizer.word_delimiter_token_id, 98]
# fmt: on
UpperCamelCase_ = tokenizer.decode(_UpperCAmelCase , output_char_offsets=_UpperCAmelCase , filter_word_delimiter_token=_UpperCAmelCase )
# check Wav2Vec2CTCTokenizerOutput keys for char
self.assertEqual(len(outputs.keys() ) , 2 )
self.assertTrue('text' in outputs )
self.assertTrue('char_offsets' in outputs )
self.assertTrue(isinstance(_UpperCAmelCase , _UpperCAmelCase ) )
# check that order of chars is correct and identical for both outputs
self.assertEqual(' '.join(self.get_from_offsets(outputs['char_offsets'] , 'char' ) ) , outputs.text )
self.assertListEqual(
self.get_from_offsets(outputs['char_offsets'] , 'char' ) , ['k', 's', 'ɾ', 'ɾ', '|', 'ɾ', 'l', '|', 'ɭʲ'] )
# check that offsets are actually correct for char
# 0-1 is 11, 1-4 is 5, 4-6 is first 15, 6-7 is <pad> (thus not shown), 7-9 is second 15, 9-10 is word_delimiter_token,
# 10-11 is <pad> (thus not shown), 11-12 is third 15, 12-15 is 8, 15-16 is word_delimiter_token, 16-17 is 98
self.assertListEqual(
self.get_from_offsets(outputs['char_offsets'] , 'start_offset' ) , [0, 1, 4, 7, 9, 11, 12, 15, 16] )
self.assertListEqual(
self.get_from_offsets(outputs['char_offsets'] , 'end_offset' ) , [1, 4, 6, 9, 10, 12, 15, 16, 17] )
def _UpperCAmelCase ( self ) -> List[str]:
UpperCamelCase_ = self.get_tokenizer(word_delimiter_token='|' )
def check_list_tuples_equal(_UpperCAmelCase , _UpperCAmelCase ):
self.assertTrue(isinstance(_UpperCAmelCase , _UpperCAmelCase ) )
self.assertTrue(isinstance(outputs_list[0] , _UpperCAmelCase ) )
# transform list to ModelOutput
UpperCamelCase_ = WavaVecaPhonemeCTCTokenizerOutput(
{k: [d[k] for d in outputs_list] for k in outputs_list[0]} )
self.assertListEqual(outputs_batch['text'] , outputs_batch_a['text'] )
def recursive_check(_UpperCAmelCase , _UpperCAmelCase ):
if isinstance(_UpperCAmelCase , _UpperCAmelCase ):
[recursive_check(_UpperCAmelCase , _UpperCAmelCase ) for la, la in zip(_UpperCAmelCase , _UpperCAmelCase )]
self.assertEqual(_UpperCAmelCase , _UpperCAmelCase )
if "char_offsets" in outputs_batch:
recursive_check(outputs_batch['char_offsets'] , outputs_batch_a['char_offsets'] )
# fmt: off
UpperCamelCase_ = [
[11, 5, 15, tokenizer.pad_token_id, 15, 4, 8, 98, 32, 32, 32, 32, 4, 33, tokenizer.word_delimiter_token_id, 32, 32, 33, 34, 34],
[24, 22, 5, tokenizer.word_delimiter_token_id, tokenizer.word_delimiter_token_id, 24, 22, 22, 22, 4, 5, 77, tokenizer.pad_token_id, 22, 22, 4, 34, 34, 34, 34],
]
# fmt: on
# We assume that `decode` works as expected. All we will check now is
# the output type is correct and the output is identical to `decode`
# char
UpperCamelCase_ = tokenizer.batch_decode(_UpperCAmelCase , output_char_offsets=_UpperCAmelCase )
UpperCamelCase_ = [tokenizer.decode(_UpperCAmelCase , output_char_offsets=_UpperCAmelCase ) for ids in sample_ids]
check_list_tuples_equal(_UpperCAmelCase , _UpperCAmelCase )
@unittest.skip('Wav2Vec2PhonemeTokenizer always lower cases letters to correctly map to phonemes' )
def _UpperCAmelCase ( self ) -> List[Any]:
pass
@unittest.skip('Wav2Vec2PhonemeTokenizer always puts spaces between phonemes' )
def _UpperCAmelCase ( self ) -> List[str]:
pass
@unittest.skip('encodes to text to ids, but decodes ids to phonemes -> not possible to have internal consistency' )
def _UpperCAmelCase ( self ) -> List[Any]:
pass
@unittest.skip('Wav2Vec2PhonemeModel has no max model length => no testing' )
def _UpperCAmelCase ( self ) -> List[Any]:
pass
def _UpperCAmelCase ( self ) -> str:
UpperCamelCase_ = self.get_tokenizers(do_lower_case=_UpperCAmelCase )
for tokenizer in tokenizers:
with self.subTest(f"""{tokenizer.__class__.__name__}""" ):
UpperCamelCase_ = tokenizer.vocab_size
UpperCamelCase_ = len(_UpperCAmelCase )
self.assertNotEqual(_UpperCAmelCase , 0 )
# We usually have added tokens from the start in tests because our vocab fixtures are
# smaller than the original vocabs - let's not assert this
# self.assertEqual(vocab_size, all_size)
UpperCamelCase_ = ['aaaaa bbbbbb', 'cccccccccdddddddd']
UpperCamelCase_ = tokenizer.add_tokens(_UpperCAmelCase )
UpperCamelCase_ = tokenizer.vocab_size
UpperCamelCase_ = len(_UpperCAmelCase )
self.assertNotEqual(_UpperCAmelCase , 0 )
self.assertEqual(_UpperCAmelCase , _UpperCAmelCase )
self.assertEqual(_UpperCAmelCase , len(_UpperCAmelCase ) )
self.assertEqual(_UpperCAmelCase , all_size + len(_UpperCAmelCase ) )
UpperCamelCase_ = tokenizer.encode('aaaaa bbbbbb low cccccccccdddddddd l' , add_special_tokens=_UpperCAmelCase )
self.assertGreaterEqual(len(_UpperCAmelCase ) , 4 )
self.assertGreater(tokens[0] , tokenizer.vocab_size - 1 )
self.assertGreater(tokens[-3] , tokenizer.vocab_size - 1 )
UpperCamelCase_ = {'eos_token': '>>>>|||<||<<|<<', 'pad_token': '<<<<<|||>|>>>>|>'}
UpperCamelCase_ = tokenizer.add_special_tokens(_UpperCAmelCase )
UpperCamelCase_ = tokenizer.vocab_size
UpperCamelCase_ = len(_UpperCAmelCase )
self.assertNotEqual(_UpperCAmelCase , 0 )
self.assertEqual(_UpperCAmelCase , _UpperCAmelCase )
self.assertEqual(_UpperCAmelCase , len(_UpperCAmelCase ) )
self.assertEqual(_UpperCAmelCase , all_size_a + len(_UpperCAmelCase ) )
UpperCamelCase_ = tokenizer.encode(
'>>>>|||<||<<|<< aaaaabbbbbb low cccccccccdddddddd <<<<<|||>|>>>>|> l' , add_special_tokens=_UpperCAmelCase )
self.assertGreaterEqual(len(_UpperCAmelCase ) , 6 )
self.assertGreater(tokens[0] , tokenizer.vocab_size - 1 )
self.assertGreater(tokens[0] , tokens[1] )
self.assertGreater(tokens[-3] , tokenizer.vocab_size - 1 )
self.assertGreater(tokens[-3] , tokens[-4] )
self.assertEqual(tokens[0] , tokenizer.eos_token_id )
self.assertEqual(tokens[-3] , tokenizer.pad_token_id )
@unittest.skip('The tokenizer shouldn\'t be used to encode input IDs (except for labels), only to decode.' )
def _UpperCAmelCase ( self ) -> List[Any]:
pass
@unittest.skip('The tokenizer shouldn\'t be used to encode input IDs (except for labels), only to decode.' )
def _UpperCAmelCase ( self ) -> List[Any]:
pass
def _UpperCAmelCase ( self ) -> Optional[int]:
# The default common tokenizer tests assumes that the output of `convert_tokens_to_string` is a string which
# is not the case for Wav2Vec2PhonemeCTCTokenizer.
UpperCamelCase_ = self.get_tokenizers(fast=_UpperCAmelCase , do_lower_case=_UpperCAmelCase )
for tokenizer in tokenizers:
with self.subTest(f"""{tokenizer.__class__.__name__}""" ):
UpperCamelCase_ = ['ð', 'ɪ', 's', 'ɪ', 'z', 'ɐ', 't', 'ɛ', 'k', 's', 't']
UpperCamelCase_ = tokenizer.convert_tokens_to_string(_UpperCAmelCase )
self.assertIsInstance(output['text'] , _UpperCAmelCase )
| 23 |
from __future__ import annotations
import sys
from collections import deque
from typing import Generic, TypeVar
snake_case__ : Dict = TypeVar("""T""")
class _a ( Generic[T] ):
"""simple docstring"""
A_ = 42 # Cache store of keys
A_ = 42 # References of the keys in cache
A_ = 10 # Maximum capacity of cache
def __init__( self , _UpperCAmelCase ) -> None:
UpperCamelCase_ = deque()
UpperCamelCase_ = set()
if not n:
UpperCamelCase_ = sys.maxsize
elif n < 0:
raise ValueError('n should be an integer greater than 0.' )
else:
UpperCamelCase_ = n
def _UpperCAmelCase ( self , _UpperCAmelCase ) -> None:
if x not in self.key_reference:
if len(self.dq_store ) == LRUCache._MAX_CAPACITY:
UpperCamelCase_ = self.dq_store.pop()
self.key_reference.remove(_UpperCAmelCase )
else:
self.dq_store.remove(_UpperCAmelCase )
self.dq_store.appendleft(_UpperCAmelCase )
self.key_reference.add(_UpperCAmelCase )
def _UpperCAmelCase ( self ) -> None:
for k in self.dq_store:
print(_UpperCAmelCase )
def __repr__( self ) -> str:
return f"""LRUCache({self._MAX_CAPACITY}) => {list(self.dq_store )}"""
if __name__ == "__main__":
import doctest
doctest.testmod()
snake_case__ : LRUCache[str | int] = LRUCache(4)
lru_cache.refer("""A""")
lru_cache.refer(2)
lru_cache.refer(3)
lru_cache.refer("""A""")
lru_cache.refer(4)
lru_cache.refer(5)
lru_cache.display()
print(lru_cache)
assert str(lru_cache) == "LRUCache(4) => [5, 4, 'A', 3]"
| 23 | 1 |
def _snake_case (__lowercase):
UpperCamelCase_ = False
while is_sorted is False: # Until all the indices are traversed keep looping
UpperCamelCase_ = True
for i in range(0 , len(__lowercase) - 1 , 2): # iterating over all even indices
if input_list[i] > input_list[i + 1]:
UpperCamelCase_ , UpperCamelCase_ = input_list[i + 1], input_list[i]
# swapping if elements not in order
UpperCamelCase_ = False
for i in range(1 , len(__lowercase) - 1 , 2): # iterating over all odd indices
if input_list[i] > input_list[i + 1]:
UpperCamelCase_ , UpperCamelCase_ = input_list[i + 1], input_list[i]
# swapping if elements not in order
UpperCamelCase_ = False
return input_list
if __name__ == "__main__":
print("""Enter list to be sorted""")
snake_case__ : Union[str, Any] = [int(x) for x in input().split()]
# inputing elements of the list in one line
snake_case__ : Dict = odd_even_sort(input_list)
print("""The sorted list is""")
print(sorted_list)
| 23 |
import numpy as np
def _snake_case (__lowercase):
return 1 / (1 + np.exp(-vector))
def _snake_case (__lowercase):
return vector * sigmoid(__lowercase)
if __name__ == "__main__":
import doctest
doctest.testmod()
| 23 | 1 |
from collections import OrderedDict
from typing import Mapping
from packaging import version
from ...configuration_utils import PretrainedConfig
from ...onnx import OnnxConfig
from ...utils import logging
snake_case__ : Tuple = logging.get_logger(__name__)
snake_case__ : List[str] = {
"""hustvl/yolos-small""": """https://huggingface.co/hustvl/yolos-small/resolve/main/config.json""",
# See all YOLOS models at https://huggingface.co/models?filter=yolos
}
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
A_ = """yolos"""
def __init__( self , _UpperCAmelCase=768 , _UpperCAmelCase=12 , _UpperCAmelCase=12 , _UpperCAmelCase=3072 , _UpperCAmelCase="gelu" , _UpperCAmelCase=0.0 , _UpperCAmelCase=0.0 , _UpperCAmelCase=0.0_2 , _UpperCAmelCase=1e-12 , _UpperCAmelCase=[512, 864] , _UpperCAmelCase=16 , _UpperCAmelCase=3 , _UpperCAmelCase=True , _UpperCAmelCase=100 , _UpperCAmelCase=True , _UpperCAmelCase=False , _UpperCAmelCase=1 , _UpperCAmelCase=5 , _UpperCAmelCase=2 , _UpperCAmelCase=5 , _UpperCAmelCase=2 , _UpperCAmelCase=0.1 , **_UpperCAmelCase , ) -> List[str]:
super().__init__(**_UpperCAmelCase )
UpperCamelCase_ = hidden_size
UpperCamelCase_ = num_hidden_layers
UpperCamelCase_ = num_attention_heads
UpperCamelCase_ = intermediate_size
UpperCamelCase_ = hidden_act
UpperCamelCase_ = hidden_dropout_prob
UpperCamelCase_ = attention_probs_dropout_prob
UpperCamelCase_ = initializer_range
UpperCamelCase_ = layer_norm_eps
UpperCamelCase_ = image_size
UpperCamelCase_ = patch_size
UpperCamelCase_ = num_channels
UpperCamelCase_ = qkv_bias
UpperCamelCase_ = num_detection_tokens
UpperCamelCase_ = use_mid_position_embeddings
UpperCamelCase_ = auxiliary_loss
# Hungarian matcher
UpperCamelCase_ = class_cost
UpperCamelCase_ = bbox_cost
UpperCamelCase_ = giou_cost
# Loss coefficients
UpperCamelCase_ = bbox_loss_coefficient
UpperCamelCase_ = giou_loss_coefficient
UpperCamelCase_ = eos_coefficient
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
A_ = version.parse("""1.11""" )
@property
def _UpperCAmelCase ( self ) -> Mapping[str, Mapping[int, str]]:
return OrderedDict(
[
('pixel_values', {0: 'batch', 1: 'num_channels', 2: 'height', 3: 'width'}),
] )
@property
def _UpperCAmelCase ( self ) -> float:
return 1e-4
@property
def _UpperCAmelCase ( self ) -> int:
return 12
| 23 |
import math
from datetime import datetime, timedelta
def _snake_case (__lowercase):
UpperCamelCase_ = year % 19
UpperCamelCase_ = year % 4
UpperCamelCase_ = year % 7
UpperCamelCase_ = math.floor(year / 100)
UpperCamelCase_ = math.floor((13 + 8 * leap_day_inhibits) / 25)
UpperCamelCase_ = leap_day_inhibits / 4
UpperCamelCase_ = (
15 - lunar_orbit_correction + leap_day_inhibits - leap_day_reinstall_number
) % 30
UpperCamelCase_ = (4 + leap_day_inhibits - leap_day_reinstall_number) % 7
# days to be added to March 21
UpperCamelCase_ = (19 * metonic_cycle + secular_moon_shift) % 30
# PHM -> Paschal Full Moon
UpperCamelCase_ = (
2 * julian_leap_year
+ 4 * non_leap_year
+ 6 * days_to_add
+ century_starting_point
) % 7
if days_to_add == 29 and days_from_phm_to_sunday == 6:
return datetime(__lowercase , 4 , 19)
elif days_to_add == 28 and days_from_phm_to_sunday == 6:
return datetime(__lowercase , 4 , 18)
else:
return datetime(__lowercase , 3 , 22) + timedelta(
days=int(days_to_add + days_from_phm_to_sunday))
if __name__ == "__main__":
for year in (1_9_9_4, 2_0_0_0, 2_0_1_0, 2_0_2_1, 2_0_2_3):
snake_case__ : Dict = """will be""" if year > datetime.now().year else """was"""
print(f'Easter in {year} {tense} {gauss_easter(year)}')
| 23 | 1 |
from sympy import diff, lambdify, symbols
from sympy.functions import * # noqa: F403
def _snake_case (__lowercase , __lowercase , __lowercase = "x" , __lowercase = 10**-10 , __lowercase = 1 , ):
UpperCamelCase_ = symbols(__lowercase)
UpperCamelCase_ = lambdify(__lowercase , __lowercase)
UpperCamelCase_ = lambdify(__lowercase , diff(__lowercase , __lowercase))
UpperCamelCase_ = starting_point
while True:
if diff_function(__lowercase) != 0:
UpperCamelCase_ = prev_guess - multiplicity * func(__lowercase) / diff_function(
__lowercase)
else:
raise ZeroDivisionError('Could not find root') from None
# Precision is checked by comparing the difference of consecutive guesses
if abs(next_guess - prev_guess) < precision:
return next_guess
UpperCamelCase_ = next_guess
# Let's Execute
if __name__ == "__main__":
# Find root of trigonometric function
# Find value of pi
print(f'The root of sin(x) = 0 is {newton_raphson("sin(x)", 2)}')
# Find root of polynomial
# Find fourth Root of 5
print(f'The root of x**4 - 5 = 0 is {newton_raphson("x**4 -5", 0.4 +5J)}')
# Find value of e
print(
"""The root of log(y) - 1 = 0 is """,
f'{newton_raphson("log(y) - 1", 2, variable="y")}',
)
# Exponential Roots
print(
"""The root of exp(x) - 1 = 0 is""",
f'{newton_raphson("exp(x) - 1", 1_0, precision=0.005)}',
)
# Find root of cos(x)
print(f'The root of cos(x) = 0 is {newton_raphson("cos(x)", 0)}')
| 23 |
import requests
def _snake_case (__lowercase , __lowercase):
UpperCamelCase_ = {'Content-Type': 'application/json'}
UpperCamelCase_ = requests.post(__lowercase , json={'text': message_body} , headers=__lowercase)
if response.status_code != 200:
UpperCamelCase_ = (
'Request to slack returned an error '
f"""{response.status_code}, the response is:\n{response.text}"""
)
raise ValueError(__lowercase)
if __name__ == "__main__":
# Set the slack url to the one provided by Slack when you create the webhook at
# https://my.slack.com/services/new/incoming-webhook/
send_slack_message("""<YOUR MESSAGE BODY>""", """<SLACK CHANNEL URL>""")
| 23 | 1 |
# Copyright 2021 The HuggingFace Team. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import warnings
from typing import List
from unittest.mock import Mock
import torch
from torch.utils.data import DataLoader, IterableDataset, TensorDataset
from accelerate.accelerator import Accelerator
from accelerate.utils.dataclasses import DistributedType
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
def __init__( self , _UpperCAmelCase ) -> Tuple:
UpperCamelCase_ = data
def __iter__( self ) -> Tuple:
for element in self.data:
yield element
def _snake_case (__lowercase=True):
UpperCamelCase_ = Accelerator(even_batches=__lowercase)
assert accelerator.num_processes == 2, "this script expects that two GPUs are available"
return accelerator
def _snake_case (__lowercase , __lowercase , __lowercase , __lowercase = False):
if iterable:
UpperCamelCase_ = DummyIterableDataset(torch.as_tensor(range(__lowercase)))
else:
UpperCamelCase_ = TensorDataset(torch.as_tensor(range(__lowercase)))
UpperCamelCase_ = DataLoader(__lowercase , batch_size=__lowercase)
UpperCamelCase_ = accelerator.prepare(__lowercase)
return dl
def _snake_case (__lowercase , __lowercase , __lowercase , __lowercase , __lowercase , ):
UpperCamelCase_ = create_dataloader(accelerator=__lowercase , dataset_size=__lowercase , batch_size=__lowercase)
UpperCamelCase_ = [len(batch[0]) for batch in dl]
if accelerator.process_index == 0:
assert batch_sizes == process_0_expected_batch_sizes
elif accelerator.process_index == 1:
assert batch_sizes == process_1_expected_batch_sizes
def _snake_case ():
UpperCamelCase_ = create_accelerator()
# without padding, we would expect a different number of batches
verify_dataloader_batch_sizes(
__lowercase , dataset_size=3 , batch_size=1 , process_0_expected_batch_sizes=[1, 1] , process_1_expected_batch_sizes=[1, 1] , )
# without padding, we would expect the same number of batches, but different sizes
verify_dataloader_batch_sizes(
__lowercase , dataset_size=7 , batch_size=2 , process_0_expected_batch_sizes=[2, 2] , process_1_expected_batch_sizes=[2, 2] , )
def _snake_case ():
UpperCamelCase_ = create_accelerator(even_batches=__lowercase)
verify_dataloader_batch_sizes(
__lowercase , dataset_size=3 , batch_size=1 , process_0_expected_batch_sizes=[1, 1] , process_1_expected_batch_sizes=[1] , )
verify_dataloader_batch_sizes(
__lowercase , dataset_size=7 , batch_size=2 , process_0_expected_batch_sizes=[2, 2] , process_1_expected_batch_sizes=[2, 1] , )
def _snake_case ():
UpperCamelCase_ = create_accelerator(even_batches=__lowercase)
UpperCamelCase_ = torch.nn.Linear(1 , 1)
UpperCamelCase_ = accelerator.prepare(__lowercase)
UpperCamelCase_ = create_dataloader(__lowercase , dataset_size=3 , batch_size=1)
UpperCamelCase_ = []
with accelerator.join_uneven_inputs([ddp_model]):
for batch_idx, batch in enumerate(__lowercase):
UpperCamelCase_ = ddp_model(batch[0].float())
UpperCamelCase_ = output.sum()
loss.backward()
batch_idxs.append(__lowercase)
accelerator.wait_for_everyone()
if accelerator.process_index == 0:
assert batch_idxs == [0, 1]
elif accelerator.process_index == 1:
assert batch_idxs == [0]
def _snake_case (__lowercase):
with warnings.catch_warnings(record=__lowercase) as w:
with accelerator.join_uneven_inputs([Mock()]):
pass
assert issubclass(w[-1].category , __lowercase)
assert "only supported for multi-GPU" in str(w[-1].message)
def _snake_case ():
UpperCamelCase_ = True
UpperCamelCase_ = False
UpperCamelCase_ = create_accelerator(even_batches=__lowercase)
UpperCamelCase_ = torch.nn.Linear(1 , 1)
UpperCamelCase_ = accelerator.prepare(__lowercase)
UpperCamelCase_ = create_dataloader(__lowercase , dataset_size=3 , batch_size=1)
UpperCamelCase_ = create_dataloader(__lowercase , dataset_size=3 , batch_size=1)
with accelerator.join_uneven_inputs([ddp_model] , even_batches=__lowercase):
UpperCamelCase_ = train_dl.batch_sampler.even_batches
UpperCamelCase_ = valid_dl.batch_sampler.even_batches
assert train_dl_overridden_value == overridden_even_batches
assert valid_dl_overridden_value == overridden_even_batches
assert train_dl.batch_sampler.even_batches == default_even_batches
assert valid_dl.batch_sampler.even_batches == default_even_batches
def _snake_case ():
UpperCamelCase_ = True
UpperCamelCase_ = False
UpperCamelCase_ = create_accelerator(even_batches=__lowercase)
UpperCamelCase_ = torch.nn.Linear(1 , 1)
UpperCamelCase_ = accelerator.prepare(__lowercase)
create_dataloader(__lowercase , dataset_size=3 , batch_size=1 , iterable=__lowercase)
UpperCamelCase_ = create_dataloader(__lowercase , dataset_size=3 , batch_size=1)
with warnings.catch_warnings():
warnings.filterwarnings('ignore')
try:
with accelerator.join_uneven_inputs([ddp_model] , even_batches=__lowercase):
UpperCamelCase_ = batch_dl.batch_sampler.even_batches
except AttributeError:
# ensure attribute error is not raised when processing iterable dl
raise AssertionError
assert batch_dl_overridden_value == overridden_even_batches
assert batch_dl.batch_sampler.even_batches == default_even_batches
def _snake_case ():
UpperCamelCase_ = create_accelerator()
UpperCamelCase_ = torch.nn.Linear(1 , 1)
UpperCamelCase_ = accelerator.prepare(__lowercase)
create_dataloader(__lowercase , dataset_size=3 , batch_size=1 , iterable=__lowercase)
with warnings.catch_warnings(record=__lowercase) as w:
with accelerator.join_uneven_inputs([ddp_model] , even_batches=__lowercase):
pass
assert issubclass(w[-1].category , __lowercase)
assert "only supported for map-style datasets" in str(w[-1].message)
def _snake_case ():
UpperCamelCase_ = create_accelerator()
accelerator.print('Test that even_batches variable ensures uniform batches across processes')
test_default_ensures_even_batch_sizes()
accelerator.print('Run tests with even_batches disabled')
test_can_disable_even_batches()
accelerator.print('Test joining uneven inputs')
test_can_join_uneven_inputs()
accelerator.print('Test overriding even_batches when joining uneven inputs')
test_join_can_override_even_batches()
accelerator.print('Test overriding even_batches for mixed dataloader types')
test_join_can_override_for_mixed_type_dataloaders()
accelerator.print('Test overriding even_batches raises a warning for iterable dataloaders')
test_join_raises_warning_for_iterable_when_overriding_even_batches()
accelerator.print('Test join with non DDP distributed raises warning')
UpperCamelCase_ = accelerator.state.distributed_type
UpperCamelCase_ = DistributedType.FSDP
test_join_raises_warning_for_non_ddp_distributed(__lowercase)
UpperCamelCase_ = original_state
if __name__ == "__main__":
main()
| 23 |
import re
from pathlib import Path
from unittest import TestCase
import pytest
@pytest.mark.integration
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
def _UpperCAmelCase ( self , _UpperCAmelCase ) -> Dict:
with open(_UpperCAmelCase , encoding='utf-8' ) as input_file:
UpperCamelCase_ = re.compile(R'(?!.*\b(?:encoding|rb|w|wb|w+|wb+|ab|ab+)\b)(?<=\s)(open)\((.*)\)' )
UpperCamelCase_ = input_file.read()
UpperCamelCase_ = regexp.search(_UpperCAmelCase )
return match
def _UpperCAmelCase ( self , _UpperCAmelCase ) -> Dict:
with open(_UpperCAmelCase , encoding='utf-8' ) as input_file:
UpperCamelCase_ = re.compile(R'#[^\r\n]*print\(|\"[^\r\n]*print\(|\"\"\".*?print\(.*?\"\"\"|(print\()' , re.DOTALL )
UpperCamelCase_ = input_file.read()
# use `re.finditer` to handle the case where the ignored groups would be matched first by `re.search`
UpperCamelCase_ = regexp.finditer(_UpperCAmelCase )
UpperCamelCase_ = [match for match in matches if match is not None and match.group(1 ) is not None]
return matches[0] if matches else None
def _UpperCAmelCase ( self ) -> List[str]:
UpperCamelCase_ = Path('./datasets' )
UpperCamelCase_ = list(dataset_paths.absolute().glob('**/*.py' ) )
for dataset in dataset_files:
if self._no_encoding_on_file_open(str(_UpperCAmelCase ) ):
raise AssertionError(f"""open(...) must use utf-8 encoding in {dataset}""" )
def _UpperCAmelCase ( self ) -> Optional[int]:
UpperCamelCase_ = Path('./datasets' )
UpperCamelCase_ = list(dataset_paths.absolute().glob('**/*.py' ) )
for dataset in dataset_files:
if self._no_print_statements(str(_UpperCAmelCase ) ):
raise AssertionError(f"""print statement found in {dataset}. Use datasets.logger/logging instead.""" )
| 23 | 1 |
import json
from typing import List, Optional, Tuple
from tokenizers import normalizers
from ...tokenization_utils_fast import PreTrainedTokenizerFast
from ...utils import logging
from .tokenization_distilbert import DistilBertTokenizer
snake_case__ : Dict = logging.get_logger(__name__)
snake_case__ : Any = {"""vocab_file""": """vocab.txt""", """tokenizer_file""": """tokenizer.json"""}
snake_case__ : List[str] = {
"""vocab_file""": {
"""distilbert-base-uncased""": """https://huggingface.co/distilbert-base-uncased/resolve/main/vocab.txt""",
"""distilbert-base-uncased-distilled-squad""": (
"""https://huggingface.co/distilbert-base-uncased-distilled-squad/resolve/main/vocab.txt"""
),
"""distilbert-base-cased""": """https://huggingface.co/distilbert-base-cased/resolve/main/vocab.txt""",
"""distilbert-base-cased-distilled-squad""": (
"""https://huggingface.co/distilbert-base-cased-distilled-squad/resolve/main/vocab.txt"""
),
"""distilbert-base-german-cased""": """https://huggingface.co/distilbert-base-german-cased/resolve/main/vocab.txt""",
"""distilbert-base-multilingual-cased""": (
"""https://huggingface.co/distilbert-base-multilingual-cased/resolve/main/vocab.txt"""
),
},
"""tokenizer_file""": {
"""distilbert-base-uncased""": """https://huggingface.co/distilbert-base-uncased/resolve/main/tokenizer.json""",
"""distilbert-base-uncased-distilled-squad""": (
"""https://huggingface.co/distilbert-base-uncased-distilled-squad/resolve/main/tokenizer.json"""
),
"""distilbert-base-cased""": """https://huggingface.co/distilbert-base-cased/resolve/main/tokenizer.json""",
"""distilbert-base-cased-distilled-squad""": (
"""https://huggingface.co/distilbert-base-cased-distilled-squad/resolve/main/tokenizer.json"""
),
"""distilbert-base-german-cased""": (
"""https://huggingface.co/distilbert-base-german-cased/resolve/main/tokenizer.json"""
),
"""distilbert-base-multilingual-cased""": (
"""https://huggingface.co/distilbert-base-multilingual-cased/resolve/main/tokenizer.json"""
),
},
}
snake_case__ : List[str] = {
"""distilbert-base-uncased""": 5_1_2,
"""distilbert-base-uncased-distilled-squad""": 5_1_2,
"""distilbert-base-cased""": 5_1_2,
"""distilbert-base-cased-distilled-squad""": 5_1_2,
"""distilbert-base-german-cased""": 5_1_2,
"""distilbert-base-multilingual-cased""": 5_1_2,
}
snake_case__ : Optional[Any] = {
"""distilbert-base-uncased""": {"""do_lower_case""": True},
"""distilbert-base-uncased-distilled-squad""": {"""do_lower_case""": True},
"""distilbert-base-cased""": {"""do_lower_case""": False},
"""distilbert-base-cased-distilled-squad""": {"""do_lower_case""": False},
"""distilbert-base-german-cased""": {"""do_lower_case""": False},
"""distilbert-base-multilingual-cased""": {"""do_lower_case""": False},
}
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
A_ = VOCAB_FILES_NAMES
A_ = PRETRAINED_VOCAB_FILES_MAP
A_ = PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES
A_ = PRETRAINED_INIT_CONFIGURATION
A_ = ["""input_ids""", """attention_mask"""]
A_ = DistilBertTokenizer
def __init__( self , _UpperCAmelCase=None , _UpperCAmelCase=None , _UpperCAmelCase=True , _UpperCAmelCase="[UNK]" , _UpperCAmelCase="[SEP]" , _UpperCAmelCase="[PAD]" , _UpperCAmelCase="[CLS]" , _UpperCAmelCase="[MASK]" , _UpperCAmelCase=True , _UpperCAmelCase=None , **_UpperCAmelCase , ) -> Optional[Any]:
super().__init__(
_UpperCAmelCase , tokenizer_file=_UpperCAmelCase , do_lower_case=_UpperCAmelCase , unk_token=_UpperCAmelCase , sep_token=_UpperCAmelCase , pad_token=_UpperCAmelCase , cls_token=_UpperCAmelCase , mask_token=_UpperCAmelCase , tokenize_chinese_chars=_UpperCAmelCase , strip_accents=_UpperCAmelCase , **_UpperCAmelCase , )
UpperCamelCase_ = json.loads(self.backend_tokenizer.normalizer.__getstate__() )
if (
normalizer_state.get('lowercase' , _UpperCAmelCase ) != do_lower_case
or normalizer_state.get('strip_accents' , _UpperCAmelCase ) != strip_accents
or normalizer_state.get('handle_chinese_chars' , _UpperCAmelCase ) != tokenize_chinese_chars
):
UpperCamelCase_ = getattr(_UpperCAmelCase , normalizer_state.pop('type' ) )
UpperCamelCase_ = do_lower_case
UpperCamelCase_ = strip_accents
UpperCamelCase_ = tokenize_chinese_chars
UpperCamelCase_ = normalizer_class(**_UpperCAmelCase )
UpperCamelCase_ = do_lower_case
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase=None ) -> str:
UpperCamelCase_ = [self.cls_token_id] + token_ids_a + [self.sep_token_id]
if token_ids_a:
output += token_ids_a + [self.sep_token_id]
return output
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase = None ) -> List[int]:
UpperCamelCase_ = [self.sep_token_id]
UpperCamelCase_ = [self.cls_token_id]
if token_ids_a is None:
return len(cls + token_ids_a + sep ) * [0]
return len(cls + token_ids_a + sep ) * [0] + len(token_ids_a + sep ) * [1]
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase = None ) -> Tuple[str]:
UpperCamelCase_ = self._tokenizer.model.save(_UpperCAmelCase , name=_UpperCAmelCase )
return tuple(_UpperCAmelCase )
| 23 |
import argparse
import random
import joblib
import numpy as np
import torch
from igf.igf import (
SecondaryLearner,
collect_objective_set,
compute_perplexity,
generate_datasets,
load_gpta,
recopy_gpta,
set_seed,
train_secondary_learner,
)
from torch.utils.data import DataLoader, RandomSampler
from transformers import GPTaLMHeadModel
def _snake_case (__lowercase=32 , __lowercase=10 , __lowercase=100 , __lowercase=1026 , __lowercase=True , __lowercase="data/tokenized_stories_train_wikitext103.jbl" , __lowercase="igf_context_pairs.jbl" , ):
set_seed(3)
# generate train_data and objective_set
UpperCamelCase_ , UpperCamelCase_ = generate_datasets(
__lowercase , __lowercase , number=__lowercase , min_len=1026 , trim=__lowercase)
# keeps model same across runs
set_seed(4)
# model, lm_optimizer, lm_scheduler = recopy_gpt2(model, device, max_steps) # store original model weights
# can we train on GPU?
UpperCamelCase_ = torch.device('cuda:0' if torch.cuda.is_available() else 'cpu')
# load pretrained model
UpperCamelCase_ = load_gpta('gpt2').to(__lowercase)
print('computing perplexity on objective set')
UpperCamelCase_ = compute_perplexity(__lowercase , __lowercase , __lowercase).item()
print('perplexity on objective set:' , __lowercase)
# collect igf pairs and save to file demo.jbl
collect_objective_set(__lowercase , __lowercase , __lowercase , __lowercase , __lowercase , __lowercase , __lowercase , __lowercase)
# clean up, delete model and data we don't need anymore
del model, train_data, objective_set
torch.cuda.empty_cache()
def _snake_case (__lowercase , __lowercase=15 , __lowercase=128 , __lowercase=100 , __lowercase="igf_model.pt" , ):
set_seed(42)
# Load pre-trained model
UpperCamelCase_ = GPTaLMHeadModel.from_pretrained('gpt2')
# Initialize secondary learner to use embedding weights of model
UpperCamelCase_ = SecondaryLearner(__lowercase)
# Train secondary learner
UpperCamelCase_ = train_secondary_learner(
__lowercase , __lowercase , max_epochs=__lowercase , batch_size=__lowercase , eval_freq=100 , igf_model_path=__lowercase , )
del model, secondary_learner_train_data
torch.cuda.empty_cache()
return secondary_learner
def _snake_case (__lowercase , __lowercase , __lowercase , __lowercase=32 , __lowercase=1000 , __lowercase=16 , __lowercase=1.0 , __lowercase=recopy_gpta , __lowercase=None , __lowercase=10 , __lowercase="gpt2_finetuned.pt" , ):
UpperCamelCase_ = torch.device('cuda:0' if torch.cuda.is_available() else 'cpu')
UpperCamelCase_ = RandomSampler(__lowercase)
UpperCamelCase_ = DataLoader(__lowercase , sampler=__lowercase)
UpperCamelCase_ = max_steps // (len(__lowercase)) + 1
UpperCamelCase_ = 0
UpperCamelCase_ = torch.zeros((1, context_len) , dtype=torch.long , device=__lowercase)
UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ = recopy_model(__lowercase , __lowercase , __lowercase)
model.train()
if secondary_learner is not None:
secondary_learner.to(__lowercase)
secondary_learner.eval()
UpperCamelCase_ = []
UpperCamelCase_ = 0
UpperCamelCase_ = []
UpperCamelCase_ = []
# Compute the performance of the transformer model at the beginning
UpperCamelCase_ = compute_perplexity(__lowercase , __lowercase , __lowercase)
test_perps.append(__lowercase)
print('Test perplexity, step' , __lowercase , ':' , __lowercase)
for epoch in range(int(__lowercase)):
for step, example in enumerate(__lowercase):
torch.cuda.empty_cache()
UpperCamelCase_ = random.randint(0 , example.size(2) - context_len - 1)
UpperCamelCase_ = example[0, 0, start : start + context_len]
lm_optimizer.zero_grad()
UpperCamelCase_ = model(__lowercase , labels=__lowercase)
UpperCamelCase_ = True
if secondary_learner is not None:
UpperCamelCase_ = secondary_learner.forward(
torch.tensor(__lowercase , dtype=torch.long , device=__lowercase).unsqueeze(0))[0].item()
observed_qs.append(float(__lowercase))
# Here we implement the simple non-constant threshold for the predicted IG(X) value
# We will decay the selectivity of our secondary learner filter from
# 1 standard deviation above average to 1 below average after 10 batches.
if global_step == 10:
UpperCamelCase_ = -1
if predicted_q < threshold:
UpperCamelCase_ = False
# If we passed the filter, add the context to the batch!
if do_backprop:
contexts.append(np.array(context.cpu()))
UpperCamelCase_ = outputs[0]
lm_loss.backward()
examples += 1
del outputs
# Once the batch is filled with enough contexts, backprop on the batch.
if examples == batch_size:
torch.cuda.empty_cache()
UpperCamelCase_ = 0
# Do LM backprop
torch.nn.utils.clip_grad_norm_(model.parameters() , 3.0)
lm_optimizer.step()
lm_scheduler.step() # Update learning rate schedule
global_step += 1
# Compute the performance of the transformer model at this batch
if global_step % eval_interval == 0:
UpperCamelCase_ = compute_perplexity(__lowercase , __lowercase , __lowercase)
test_perps.append(__lowercase)
print('Test perplexity, step' , __lowercase , ':' , __lowercase)
# Break out of the loop after 60 batches
if max_steps > 0 and global_step > 60:
break
if max_steps > 0 and global_step > 60:
break
# save finetuned transformer model
torch.save(model.state_dict() , __lowercase)
torch.cuda.empty_cache()
# Do some cleaning up so we can reinitialize for the next run of this function
del lm_optimizer
del lm_scheduler
return model
def _snake_case ():
UpperCamelCase_ = argparse.ArgumentParser(description='Fine-tune a transformer model with IGF on a language modeling task')
# Required parameters
parser.add_argument(
'--data_dir' , default=__lowercase , type=__lowercase , required=__lowercase , help='The input data dir. Should contain data files for WikiText.' , )
parser.add_argument(
'--model_name_or_path' , default=__lowercase , type=__lowercase , required=__lowercase , help='Path to pretrained model or model identifier from huggingface.co/models' , )
parser.add_argument(
'--data_file' , type=__lowercase , default=__lowercase , help=(
'A jbl file containing tokenized data which can be split as objective dataset, '
'train_dataset and test_dataset.'
) , )
parser.add_argument(
'--igf_data_file' , type=__lowercase , default=__lowercase , help='A jbl file containing the context and information gain pairs to train secondary learner.' , )
parser.add_argument(
'--output_dir' , default=__lowercase , type=__lowercase , required=__lowercase , help='The output directory where the final fine-tuned model is stored.' , )
parser.add_argument(
'--tokenizer_name' , default=__lowercase , type=__lowercase , help='Pretrained tokenizer name or path if not the same as model_name' , )
parser.add_argument('--seed' , type=__lowercase , default=__lowercase , help='A seed for reproducible training.')
parser.add_argument(
'--context_len' , default=32 , type=__lowercase , help=(
'The maximum total input sequence length after tokenization. Sequences longer '
'than this will be truncated, sequences shorter will be padded.'
) , )
parser.add_argument(
'--size_objective_set' , default=100 , type=__lowercase , help='number of articles that are long enough to be used as our objective set' , )
parser.add_argument(
'--eval_freq' , default=100 , type=__lowercase , help='secondary model evaluation is triggered at eval_freq')
parser.add_argument('--max_steps' , default=1000 , type=__lowercase , help='To calculate training epochs')
parser.add_argument(
'--secondary_learner_batch_size' , default=128 , type=__lowercase , help='batch size of training data for secondary learner' , )
parser.add_argument(
'--batch_size' , default=16 , type=__lowercase , help='batch size of training data of language model(gpt2) ')
parser.add_argument(
'--eval_interval' , default=10 , type=__lowercase , help=(
'decay the selectivity of our secondary learner filter from'
'1 standard deviation above average to 1 below average after 10 batches'
) , )
parser.add_argument(
'--number' , default=100 , type=__lowercase , help='The number of examples split to be used as objective_set/test_data')
parser.add_argument(
'--min_len' , default=1026 , type=__lowercase , help='The minimum length of the article to be used as objective set')
parser.add_argument(
'--secondary_learner_max_epochs' , default=15 , type=__lowercase , help='number of epochs to train secondary learner')
parser.add_argument('--trim' , default=__lowercase , type=__lowercase , help='truncate the example if it exceeds context length')
parser.add_argument(
'--threshold' , default=1.0 , type=__lowercase , help=(
'The threshold value used by secondary learner to filter the train_data and allow only'
' informative data as input to the model'
) , )
parser.add_argument('--finetuned_model_name' , default='gpt2_finetuned.pt' , type=__lowercase , help='finetuned_model_name')
parser.add_argument(
'--recopy_model' , default=__lowercase , type=__lowercase , help='Reset the model to the original pretrained GPT-2 weights after each iteration' , )
# function calls
# Collecting *n* pairs of context and information gain(X, IG(X)) for training the secondary learner
generate_n_pairs(
context_len=32 , max_steps=10 , size_objective_set=100 , min_len=1026 , trim=__lowercase , data_file='data/tokenized_stories_train_wikitext103.jbl' , igf_data_file='igf_context_pairs.jbl' , )
# Load train data for secondary learner
UpperCamelCase_ = joblib.load('data/IGF_values.jbl')
# Train secondary learner
UpperCamelCase_ = training_secondary_learner(
__lowercase , secondary_learner_max_epochs=15 , secondary_learner_batch_size=128 , eval_freq=100 , igf_model_path='igf_model.pt' , )
# load pretrained gpt2 model
UpperCamelCase_ = GPTaLMHeadModel.from_pretrained('gpt2')
set_seed(42)
# Generate train and test data to train and evaluate gpt2 model
UpperCamelCase_ , UpperCamelCase_ = generate_datasets(
context_len=32 , file='data/tokenized_stories_train_wikitext103.jbl' , number=100 , min_len=1026 , trim=__lowercase)
# fine-tuning of the gpt2 model using igf (Information Gain Filtration)
finetune(
__lowercase , __lowercase , __lowercase , context_len=32 , max_steps=1000 , batch_size=16 , threshold=1.0 , recopy_model=__lowercase , secondary_learner=__lowercase , eval_interval=10 , finetuned_model_name='gpt2_finetuned.pt' , )
if __name__ == "__main__":
main()
| 23 | 1 |
from collections import defaultdict
from typing import Optional
from ..image_utils import load_image
from ..utils import (
add_end_docstrings,
is_torch_available,
logging,
requires_backends,
)
from .base import PIPELINE_INIT_ARGS, ChunkPipeline
if is_torch_available():
import torch
from ..models.auto.modeling_auto import MODEL_FOR_MASK_GENERATION_MAPPING
snake_case__ : str = logging.get_logger(__name__)
@add_end_docstrings(UpperCAmelCase__ )
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
def __init__( self , **_UpperCAmelCase ) -> Optional[Any]:
super().__init__(**_UpperCAmelCase )
requires_backends(self , 'vision' )
requires_backends(self , 'torch' )
if self.framework != "pt":
raise ValueError(f"""The {self.__class__} is only available in PyTorch.""" )
self.check_model_type(_UpperCAmelCase )
def _UpperCAmelCase ( self , **_UpperCAmelCase ) -> Any:
UpperCamelCase_ = {}
UpperCamelCase_ = {}
UpperCamelCase_ = {}
# preprocess args
if "points_per_batch" in kwargs:
UpperCamelCase_ = kwargs['points_per_batch']
if "points_per_crop" in kwargs:
UpperCamelCase_ = kwargs['points_per_crop']
if "crops_n_layers" in kwargs:
UpperCamelCase_ = kwargs['crops_n_layers']
if "crop_overlap_ratio" in kwargs:
UpperCamelCase_ = kwargs['crop_overlap_ratio']
if "crop_n_points_downscale_factor" in kwargs:
UpperCamelCase_ = kwargs['crop_n_points_downscale_factor']
# postprocess args
if "pred_iou_thresh" in kwargs:
UpperCamelCase_ = kwargs['pred_iou_thresh']
if "stability_score_offset" in kwargs:
UpperCamelCase_ = kwargs['stability_score_offset']
if "mask_threshold" in kwargs:
UpperCamelCase_ = kwargs['mask_threshold']
if "stability_score_thresh" in kwargs:
UpperCamelCase_ = kwargs['stability_score_thresh']
if "crops_nms_thresh" in kwargs:
UpperCamelCase_ = kwargs['crops_nms_thresh']
if "output_rle_mask" in kwargs:
UpperCamelCase_ = kwargs['output_rle_mask']
if "output_bboxes_mask" in kwargs:
UpperCamelCase_ = kwargs['output_bboxes_mask']
return preprocess_kwargs, forward_params, postprocess_kwargs
def __call__( self , _UpperCAmelCase , *_UpperCAmelCase , _UpperCAmelCase=None , _UpperCAmelCase=None , **_UpperCAmelCase ) -> List[str]:
return super().__call__(_UpperCAmelCase , *_UpperCAmelCase , num_workers=_UpperCAmelCase , batch_size=_UpperCAmelCase , **_UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase=64 , _UpperCAmelCase = 0 , _UpperCAmelCase = 512 / 1500 , _UpperCAmelCase = 32 , _UpperCAmelCase = 1 , ) -> Optional[Any]:
UpperCamelCase_ = load_image(_UpperCAmelCase )
UpperCamelCase_ = self.image_processor.size['longest_edge']
UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ = self.image_processor.generate_crop_boxes(
_UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase )
UpperCamelCase_ = self.image_processor(images=_UpperCAmelCase , return_tensors='pt' )
with self.device_placement():
if self.framework == "pt":
UpperCamelCase_ = self.get_inference_context()
with inference_context():
UpperCamelCase_ = self._ensure_tensor_on_device(_UpperCAmelCase , device=self.device )
UpperCamelCase_ = self.model.get_image_embeddings(model_inputs.pop('pixel_values' ) )
UpperCamelCase_ = image_embeddings
UpperCamelCase_ = grid_points.shape[1]
UpperCamelCase_ = points_per_batch if points_per_batch is not None else n_points
if points_per_batch <= 0:
raise ValueError(
'Cannot have points_per_batch<=0. Must be >=1 to returned batched outputs. '
'To return all points at once, set points_per_batch to None' )
for i in range(0 , _UpperCAmelCase , _UpperCAmelCase ):
UpperCamelCase_ = grid_points[:, i : i + points_per_batch, :, :]
UpperCamelCase_ = input_labels[:, i : i + points_per_batch]
UpperCamelCase_ = i == n_points - points_per_batch
yield {
"input_points": batched_points,
"input_labels": labels,
"input_boxes": crop_boxes,
"is_last": is_last,
**model_inputs,
}
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase=0.8_8 , _UpperCAmelCase=0.9_5 , _UpperCAmelCase=0 , _UpperCAmelCase=1 , ) -> List[Any]:
UpperCamelCase_ = model_inputs.pop('input_boxes' )
UpperCamelCase_ = model_inputs.pop('is_last' )
UpperCamelCase_ = model_inputs.pop('original_sizes' ).tolist()
UpperCamelCase_ = model_inputs.pop('reshaped_input_sizes' ).tolist()
UpperCamelCase_ = self.model(**_UpperCAmelCase )
# post processing happens here in order to avoid CPU GPU copies of ALL the masks
UpperCamelCase_ = model_outputs['pred_masks']
UpperCamelCase_ = self.image_processor.post_process_masks(
_UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , binarize=_UpperCAmelCase )
UpperCamelCase_ = model_outputs['iou_scores']
UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ = self.image_processor.filter_masks(
masks[0] , iou_scores[0] , original_sizes[0] , input_boxes[0] , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , )
return {
"masks": masks,
"is_last": is_last,
"boxes": boxes,
"iou_scores": iou_scores,
}
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase=False , _UpperCAmelCase=False , _UpperCAmelCase=0.7 , ) -> Any:
UpperCamelCase_ = []
UpperCamelCase_ = []
UpperCamelCase_ = []
for model_output in model_outputs:
all_scores.append(model_output.pop('iou_scores' ) )
all_masks.extend(model_output.pop('masks' ) )
all_boxes.append(model_output.pop('boxes' ) )
UpperCamelCase_ = torch.cat(_UpperCAmelCase )
UpperCamelCase_ = torch.cat(_UpperCAmelCase )
UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ = self.image_processor.post_process_for_mask_generation(
_UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase )
UpperCamelCase_ = defaultdict(_UpperCAmelCase )
for output in model_outputs:
for k, v in output.items():
extra[k].append(_UpperCAmelCase )
UpperCamelCase_ = {}
if output_rle_mask:
UpperCamelCase_ = rle_mask
if output_bboxes_mask:
UpperCamelCase_ = bounding_boxes
return {"masks": output_masks, "scores": iou_scores, **optional, **extra}
| 23 |
from __future__ import annotations
import unittest
from transformers import AutoTokenizer, MBartConfig, is_tf_available
from transformers.testing_utils import require_sentencepiece, require_tf, require_tokenizers, slow
from transformers.utils import cached_property
from ...test_configuration_common import ConfigTester
from ...test_modeling_tf_common import TFModelTesterMixin, ids_tensor
from ...test_pipeline_mixin import PipelineTesterMixin
if is_tf_available():
import tensorflow as tf
from transformers import TFAutoModelForSeqaSeqLM, TFMBartForConditionalGeneration, TFMBartModel
@require_tf
class _a :
"""simple docstring"""
A_ = MBartConfig
A_ = {}
A_ = """gelu"""
def __init__( self , _UpperCAmelCase , _UpperCAmelCase=13 , _UpperCAmelCase=7 , _UpperCAmelCase=True , _UpperCAmelCase=False , _UpperCAmelCase=99 , _UpperCAmelCase=32 , _UpperCAmelCase=2 , _UpperCAmelCase=4 , _UpperCAmelCase=37 , _UpperCAmelCase=0.1 , _UpperCAmelCase=0.1 , _UpperCAmelCase=20 , _UpperCAmelCase=2 , _UpperCAmelCase=1 , _UpperCAmelCase=0 , ) -> Union[str, Any]:
UpperCamelCase_ = parent
UpperCamelCase_ = batch_size
UpperCamelCase_ = seq_length
UpperCamelCase_ = is_training
UpperCamelCase_ = use_labels
UpperCamelCase_ = vocab_size
UpperCamelCase_ = hidden_size
UpperCamelCase_ = num_hidden_layers
UpperCamelCase_ = num_attention_heads
UpperCamelCase_ = intermediate_size
UpperCamelCase_ = hidden_dropout_prob
UpperCamelCase_ = attention_probs_dropout_prob
UpperCamelCase_ = max_position_embeddings
UpperCamelCase_ = eos_token_id
UpperCamelCase_ = pad_token_id
UpperCamelCase_ = bos_token_id
def _UpperCAmelCase ( self ) -> List[Any]:
UpperCamelCase_ = ids_tensor([self.batch_size, self.seq_length - 1] , self.vocab_size )
UpperCamelCase_ = tf.expand_dims(tf.constant([self.eos_token_id] * self.batch_size ) , 1 )
UpperCamelCase_ = tf.concat([input_ids, eos_tensor] , axis=1 )
UpperCamelCase_ = ids_tensor([self.batch_size, self.seq_length] , self.vocab_size )
UpperCamelCase_ = self.config_cls(
vocab_size=self.vocab_size , d_model=self.hidden_size , encoder_layers=self.num_hidden_layers , decoder_layers=self.num_hidden_layers , encoder_attention_heads=self.num_attention_heads , decoder_attention_heads=self.num_attention_heads , encoder_ffn_dim=self.intermediate_size , decoder_ffn_dim=self.intermediate_size , dropout=self.hidden_dropout_prob , attention_dropout=self.attention_probs_dropout_prob , max_position_embeddings=self.max_position_embeddings , eos_token_ids=[2] , bos_token_id=self.bos_token_id , pad_token_id=self.pad_token_id , decoder_start_token_id=self.pad_token_id , **self.config_updates , )
UpperCamelCase_ = prepare_mbart_inputs_dict(_UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase )
return config, inputs_dict
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> List[str]:
UpperCamelCase_ = TFMBartModel(config=_UpperCAmelCase ).get_decoder()
UpperCamelCase_ = inputs_dict['input_ids']
UpperCamelCase_ = input_ids[:1, :]
UpperCamelCase_ = inputs_dict['attention_mask'][:1, :]
UpperCamelCase_ = inputs_dict['head_mask']
UpperCamelCase_ = 1
# first forward pass
UpperCamelCase_ = model(_UpperCAmelCase , attention_mask=_UpperCAmelCase , head_mask=_UpperCAmelCase , use_cache=_UpperCAmelCase )
UpperCamelCase_ , UpperCamelCase_ = outputs.to_tuple()
UpperCamelCase_ = past_key_values[1]
def _snake_case (__lowercase , __lowercase , __lowercase , __lowercase=None , __lowercase=None , __lowercase=None , __lowercase=None , __lowercase=None , ):
if attention_mask is None:
UpperCamelCase_ = tf.cast(tf.math.not_equal(__lowercase , config.pad_token_id) , tf.inta)
if decoder_attention_mask is None:
UpperCamelCase_ = tf.concat(
[
tf.ones(decoder_input_ids[:, :1].shape , dtype=tf.inta),
tf.cast(tf.math.not_equal(decoder_input_ids[:, 1:] , config.pad_token_id) , tf.inta),
] , axis=-1 , )
if head_mask is None:
UpperCamelCase_ = tf.ones((config.encoder_layers, config.encoder_attention_heads))
if decoder_head_mask is None:
UpperCamelCase_ = tf.ones((config.decoder_layers, config.decoder_attention_heads))
if cross_attn_head_mask is None:
UpperCamelCase_ = tf.ones((config.decoder_layers, config.decoder_attention_heads))
return {
"input_ids": input_ids,
"decoder_input_ids": decoder_input_ids,
"attention_mask": attention_mask,
"decoder_attention_mask": decoder_attention_mask,
"head_mask": head_mask,
"decoder_head_mask": decoder_head_mask,
"cross_attn_head_mask": cross_attn_head_mask,
}
@require_tf
class _a ( UpperCAmelCase__ , UpperCAmelCase__ , unittest.TestCase ):
"""simple docstring"""
A_ = (TFMBartForConditionalGeneration, TFMBartModel) if is_tf_available() else ()
A_ = (TFMBartForConditionalGeneration,) if is_tf_available() else ()
A_ = (
{
"""conversational""": TFMBartForConditionalGeneration,
"""feature-extraction""": TFMBartModel,
"""summarization""": TFMBartForConditionalGeneration,
"""text2text-generation""": TFMBartForConditionalGeneration,
"""translation""": TFMBartForConditionalGeneration,
}
if is_tf_available()
else {}
)
A_ = True
A_ = False
A_ = False
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase ) -> Tuple:
if pipeline_test_casse_name != "FeatureExtractionPipelineTests":
# Exception encountered when calling layer '...'
return True
return False
def _UpperCAmelCase ( self ) -> Optional[int]:
UpperCamelCase_ = TFMBartModelTester(self )
UpperCamelCase_ = ConfigTester(self , config_class=_UpperCAmelCase )
def _UpperCAmelCase ( self ) -> Optional[int]:
self.config_tester.run_common_tests()
def _UpperCAmelCase ( self ) -> List[Any]:
UpperCamelCase_ = self.model_tester.prepare_config_and_inputs_for_common()
self.model_tester.check_decoder_model_past_large_inputs(*_UpperCAmelCase )
@require_sentencepiece
@require_tokenizers
@require_tf
class _a ( unittest.TestCase ):
"""simple docstring"""
A_ = [
""" UN Chief Says There Is No Military Solution in Syria""",
]
A_ = [
"""Şeful ONU declară că nu există o soluţie militară în Siria""",
]
A_ = """facebook/mbart-large-en-ro"""
@cached_property
def _UpperCAmelCase ( self ) -> Any:
return AutoTokenizer.from_pretrained(self.model_name )
@cached_property
def _UpperCAmelCase ( self ) -> List[str]:
UpperCamelCase_ = TFAutoModelForSeqaSeqLM.from_pretrained(self.model_name )
return model
def _UpperCAmelCase ( self , **_UpperCAmelCase ) -> int:
UpperCamelCase_ = self.translate_src_text(**_UpperCAmelCase )
self.assertListEqual(self.expected_text , _UpperCAmelCase )
def _UpperCAmelCase ( self , **_UpperCAmelCase ) -> List[str]:
UpperCamelCase_ = self.tokenizer(self.src_text , **_UpperCAmelCase , return_tensors='tf' )
UpperCamelCase_ = self.model.generate(
model_inputs.input_ids , attention_mask=model_inputs.attention_mask , num_beams=2 )
UpperCamelCase_ = self.tokenizer.batch_decode(_UpperCAmelCase , skip_special_tokens=_UpperCAmelCase )
return generated_words
@slow
def _UpperCAmelCase ( self ) -> List[Any]:
self._assert_generated_batch_equal_expected()
| 23 | 1 |
import copy
from typing import TYPE_CHECKING, Any, Mapping, Optional, OrderedDict
from packaging import version
from ...configuration_utils import PretrainedConfig
from ...onnx import OnnxConfig
from ...utils import logging
from ..auto.configuration_auto import AutoConfig
if TYPE_CHECKING:
from ... import PreTrainedTokenizerBase, TensorType
snake_case__ : List[str] = logging.get_logger(__name__)
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
A_ = """vision-encoder-decoder"""
A_ = True
def __init__( self , **_UpperCAmelCase ) -> Dict:
super().__init__(**_UpperCAmelCase )
if "encoder" not in kwargs or "decoder" not in kwargs:
raise ValueError(
f"""A configuraton of type {self.model_type} cannot be instantiated because """
f"""not both `encoder` and `decoder` sub-configurations are passed, but only {kwargs}""" )
UpperCamelCase_ = kwargs.pop('encoder' )
UpperCamelCase_ = encoder_config.pop('model_type' )
UpperCamelCase_ = kwargs.pop('decoder' )
UpperCamelCase_ = decoder_config.pop('model_type' )
UpperCamelCase_ = AutoConfig.for_model(_UpperCAmelCase , **_UpperCAmelCase )
UpperCamelCase_ = AutoConfig.for_model(_UpperCAmelCase , **_UpperCAmelCase )
UpperCamelCase_ = True
@classmethod
def _UpperCAmelCase ( cls , _UpperCAmelCase , _UpperCAmelCase , **_UpperCAmelCase ) -> PretrainedConfig:
logger.info('Setting `config.is_decoder=True` and `config.add_cross_attention=True` for decoder_config' )
UpperCamelCase_ = True
UpperCamelCase_ = True
return cls(encoder=encoder_config.to_dict() , decoder=decoder_config.to_dict() , **_UpperCAmelCase )
def _UpperCAmelCase ( self ) -> int:
UpperCamelCase_ = copy.deepcopy(self.__dict__ )
UpperCamelCase_ = self.encoder.to_dict()
UpperCamelCase_ = self.decoder.to_dict()
UpperCamelCase_ = self.__class__.model_type
return output
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
A_ = version.parse("""1.11""" )
@property
def _UpperCAmelCase ( self ) -> Mapping[str, Mapping[int, str]]:
return OrderedDict(
[
('pixel_values', {0: 'batch', 1: 'num_channels', 2: 'height', 3: 'width'}),
] )
@property
def _UpperCAmelCase ( self ) -> float:
return 1e-4
@property
def _UpperCAmelCase ( self ) -> Mapping[str, Mapping[int, str]]:
return OrderedDict({'last_hidden_state': {0: 'batch', 1: 'encoder_sequence'}} )
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
@property
def _UpperCAmelCase ( self ) -> Mapping[str, Mapping[int, str]]:
UpperCamelCase_ = OrderedDict()
UpperCamelCase_ = {0: 'batch', 1: 'past_decoder_sequence + sequence'}
UpperCamelCase_ = {0: 'batch', 1: 'past_decoder_sequence + sequence'}
UpperCamelCase_ = {0: 'batch', 1: 'encoder_sequence'}
return common_inputs
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase = -1 , _UpperCAmelCase = -1 , _UpperCAmelCase = False , _UpperCAmelCase = None , ) -> Mapping[str, Any]:
import torch
UpperCamelCase_ = OrderedDict()
UpperCamelCase_ = super().generate_dummy_inputs(
_UpperCAmelCase , batch_size=_UpperCAmelCase , seq_length=_UpperCAmelCase , is_pair=_UpperCAmelCase , framework=_UpperCAmelCase )
UpperCamelCase_ , UpperCamelCase_ = dummy_input['input_ids'].shape
UpperCamelCase_ = (batch, encoder_sequence, self._config.encoder_hidden_size)
UpperCamelCase_ = dummy_input.pop('input_ids' )
UpperCamelCase_ = dummy_input.pop('attention_mask' )
UpperCamelCase_ = torch.zeros(_UpperCAmelCase )
return common_inputs
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
@property
def _UpperCAmelCase ( self ) -> None:
pass
def _UpperCAmelCase ( self , _UpperCAmelCase ) -> OnnxConfig:
return VisionEncoderDecoderEncoderOnnxConfig(_UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase = "default" ) -> OnnxConfig:
UpperCamelCase_ = encoder_config.hidden_size
return VisionEncoderDecoderDecoderOnnxConfig(_UpperCAmelCase , _UpperCAmelCase )
| 23 |
def _snake_case (__lowercase):
UpperCamelCase_ = 1
for i in range(1 , num + 1):
fact *= i
return fact
def _snake_case (__lowercase):
UpperCamelCase_ = 0
while number > 0:
UpperCamelCase_ = number % 10
sum_of_digits += last_digit
UpperCamelCase_ = number // 10 # Removing the last_digit from the given number
return sum_of_digits
def _snake_case (__lowercase = 100):
UpperCamelCase_ = factorial(__lowercase)
UpperCamelCase_ = split_and_add(__lowercase)
return result
if __name__ == "__main__":
print(solution(int(input("""Enter the Number: """).strip())))
| 23 | 1 |
import numpy as np
def _snake_case (__lowercase):
return 1 / (1 + np.exp(-vector))
def _snake_case (__lowercase):
return vector * sigmoid(__lowercase)
if __name__ == "__main__":
import doctest
doctest.testmod()
| 23 |
from typing import Dict, List, Optional, Union
import numpy as np
from ...image_processing_utils import BaseImageProcessor, BatchFeature, get_size_dict
from ...image_transforms import (
center_crop,
get_resize_output_image_size,
normalize,
rescale,
resize,
to_channel_dimension_format,
)
from ...image_utils import (
IMAGENET_STANDARD_MEAN,
IMAGENET_STANDARD_STD,
ChannelDimension,
ImageInput,
PILImageResampling,
is_valid_image,
to_numpy_array,
valid_images,
)
from ...utils import TensorType, is_vision_available, logging
if is_vision_available():
import PIL
snake_case__ : str = logging.get_logger(__name__)
def _snake_case (__lowercase):
if isinstance(__lowercase , (list, tuple)) and isinstance(videos[0] , (list, tuple)) and is_valid_image(videos[0][0]):
return videos
elif isinstance(__lowercase , (list, tuple)) and is_valid_image(videos[0]):
return [videos]
elif is_valid_image(__lowercase):
return [[videos]]
raise ValueError(f"""Could not make batched video from {videos}""")
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
A_ = ["""pixel_values"""]
def __init__( self , _UpperCAmelCase = True , _UpperCAmelCase = None , _UpperCAmelCase = PILImageResampling.BILINEAR , _UpperCAmelCase = True , _UpperCAmelCase = None , _UpperCAmelCase = True , _UpperCAmelCase = 1 / 255 , _UpperCAmelCase = True , _UpperCAmelCase = None , _UpperCAmelCase = None , **_UpperCAmelCase , ) -> None:
super().__init__(**_UpperCAmelCase )
UpperCamelCase_ = size if size is not None else {'shortest_edge': 224}
UpperCamelCase_ = get_size_dict(_UpperCAmelCase , default_to_square=_UpperCAmelCase )
UpperCamelCase_ = crop_size if crop_size is not None else {'height': 224, 'width': 224}
UpperCamelCase_ = get_size_dict(_UpperCAmelCase , param_name='crop_size' )
UpperCamelCase_ = do_resize
UpperCamelCase_ = size
UpperCamelCase_ = do_center_crop
UpperCamelCase_ = crop_size
UpperCamelCase_ = resample
UpperCamelCase_ = do_rescale
UpperCamelCase_ = rescale_factor
UpperCamelCase_ = do_normalize
UpperCamelCase_ = image_mean if image_mean is not None else IMAGENET_STANDARD_MEAN
UpperCamelCase_ = image_std if image_std is not None else IMAGENET_STANDARD_STD
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase = PILImageResampling.BILINEAR , _UpperCAmelCase = None , **_UpperCAmelCase , ) -> np.ndarray:
UpperCamelCase_ = get_size_dict(_UpperCAmelCase , default_to_square=_UpperCAmelCase )
if "shortest_edge" in size:
UpperCamelCase_ = get_resize_output_image_size(_UpperCAmelCase , size['shortest_edge'] , default_to_square=_UpperCAmelCase )
elif "height" in size and "width" in size:
UpperCamelCase_ = (size['height'], size['width'])
else:
raise ValueError(f"""Size must have 'height' and 'width' or 'shortest_edge' as keys. Got {size.keys()}""" )
return resize(_UpperCAmelCase , size=_UpperCAmelCase , resample=_UpperCAmelCase , data_format=_UpperCAmelCase , **_UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase = None , **_UpperCAmelCase , ) -> np.ndarray:
UpperCamelCase_ = get_size_dict(_UpperCAmelCase )
if "height" not in size or "width" not in size:
raise ValueError(f"""Size must have 'height' and 'width' as keys. Got {size.keys()}""" )
return center_crop(_UpperCAmelCase , size=(size['height'], size['width']) , data_format=_UpperCAmelCase , **_UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase = None , **_UpperCAmelCase , ) -> int:
return rescale(_UpperCAmelCase , scale=_UpperCAmelCase , data_format=_UpperCAmelCase , **_UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase = None , **_UpperCAmelCase , ) -> np.ndarray:
return normalize(_UpperCAmelCase , mean=_UpperCAmelCase , std=_UpperCAmelCase , data_format=_UpperCAmelCase , **_UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = ChannelDimension.FIRST , ) -> np.ndarray:
if do_resize and size is None or resample is None:
raise ValueError('Size and resample must be specified if do_resize is True.' )
if do_center_crop and crop_size is None:
raise ValueError('Crop size must be specified if do_center_crop is True.' )
if do_rescale and rescale_factor is None:
raise ValueError('Rescale factor must be specified if do_rescale is True.' )
if do_normalize and (image_mean is None or image_std is None):
raise ValueError('Image mean and std must be specified if do_normalize is True.' )
# All transformations expect numpy arrays.
UpperCamelCase_ = to_numpy_array(_UpperCAmelCase )
if do_resize:
UpperCamelCase_ = self.resize(image=_UpperCAmelCase , size=_UpperCAmelCase , resample=_UpperCAmelCase )
if do_center_crop:
UpperCamelCase_ = self.center_crop(_UpperCAmelCase , size=_UpperCAmelCase )
if do_rescale:
UpperCamelCase_ = self.rescale(image=_UpperCAmelCase , scale=_UpperCAmelCase )
if do_normalize:
UpperCamelCase_ = self.normalize(image=_UpperCAmelCase , mean=_UpperCAmelCase , std=_UpperCAmelCase )
UpperCamelCase_ = to_channel_dimension_format(_UpperCAmelCase , _UpperCAmelCase )
return image
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = ChannelDimension.FIRST , **_UpperCAmelCase , ) -> PIL.Image.Image:
UpperCamelCase_ = do_resize if do_resize is not None else self.do_resize
UpperCamelCase_ = resample if resample is not None else self.resample
UpperCamelCase_ = do_center_crop if do_center_crop is not None else self.do_center_crop
UpperCamelCase_ = do_rescale if do_rescale is not None else self.do_rescale
UpperCamelCase_ = rescale_factor if rescale_factor is not None else self.rescale_factor
UpperCamelCase_ = do_normalize if do_normalize is not None else self.do_normalize
UpperCamelCase_ = image_mean if image_mean is not None else self.image_mean
UpperCamelCase_ = image_std if image_std is not None else self.image_std
UpperCamelCase_ = size if size is not None else self.size
UpperCamelCase_ = get_size_dict(_UpperCAmelCase , default_to_square=_UpperCAmelCase )
UpperCamelCase_ = crop_size if crop_size is not None else self.crop_size
UpperCamelCase_ = get_size_dict(_UpperCAmelCase , param_name='crop_size' )
if not valid_images(_UpperCAmelCase ):
raise ValueError(
'Invalid image type. Must be of type PIL.Image.Image, numpy.ndarray, '
'torch.Tensor, tf.Tensor or jax.ndarray.' )
UpperCamelCase_ = make_batched(_UpperCAmelCase )
UpperCamelCase_ = [
[
self._preprocess_image(
image=_UpperCAmelCase , do_resize=_UpperCAmelCase , size=_UpperCAmelCase , resample=_UpperCAmelCase , do_center_crop=_UpperCAmelCase , crop_size=_UpperCAmelCase , do_rescale=_UpperCAmelCase , rescale_factor=_UpperCAmelCase , do_normalize=_UpperCAmelCase , image_mean=_UpperCAmelCase , image_std=_UpperCAmelCase , data_format=_UpperCAmelCase , )
for img in video
]
for video in videos
]
UpperCamelCase_ = {'pixel_values': videos}
return BatchFeature(data=_UpperCAmelCase , tensor_type=_UpperCAmelCase )
| 23 | 1 |
def _snake_case (__lowercase , __lowercase):
UpperCamelCase_ = len(__lowercase)
UpperCamelCase_ = [[False] * (required_sum + 1) for _ in range(arr_len + 1)]
# for each arr value, a sum of zero(0) can be formed by not taking any element
# hence True/1
for i in range(arr_len + 1):
UpperCamelCase_ = True
# sum is not zero and set is empty then false
for i in range(1 , required_sum + 1):
UpperCamelCase_ = False
for i in range(1 , arr_len + 1):
for j in range(1 , required_sum + 1):
if arr[i - 1] > j:
UpperCamelCase_ = subset[i - 1][j]
if arr[i - 1] <= j:
UpperCamelCase_ = subset[i - 1][j] or subset[i - 1][j - arr[i - 1]]
return subset[arr_len][required_sum]
if __name__ == "__main__":
import doctest
doctest.testmod()
| 23 |
# DISCLAIMER: This file is strongly influenced by https://github.com/yang-song/score_sde_pytorch
import math
from dataclasses import dataclass
from typing import Optional, Tuple, Union
import torch
from ..configuration_utils import ConfigMixin, register_to_config
from ..utils import BaseOutput, randn_tensor
from .scheduling_utils import SchedulerMixin, SchedulerOutput
@dataclass
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
A_ = 42
A_ = 42
class _a ( UpperCAmelCase__ , UpperCAmelCase__ ):
"""simple docstring"""
A_ = 1
@register_to_config
def __init__( self , _UpperCAmelCase = 2000 , _UpperCAmelCase = 0.1_5 , _UpperCAmelCase = 0.0_1 , _UpperCAmelCase = 1_3_4_8.0 , _UpperCAmelCase = 1e-5 , _UpperCAmelCase = 1 , ) -> Tuple:
# standard deviation of the initial noise distribution
UpperCamelCase_ = sigma_max
# setable values
UpperCamelCase_ = None
self.set_sigmas(_UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase = None ) -> torch.FloatTensor:
return sample
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase = None , _UpperCAmelCase = None ) -> str:
UpperCamelCase_ = sampling_eps if sampling_eps is not None else self.config.sampling_eps
UpperCamelCase_ = torch.linspace(1 , _UpperCAmelCase , _UpperCAmelCase , device=_UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None ) -> Any:
UpperCamelCase_ = sigma_min if sigma_min is not None else self.config.sigma_min
UpperCamelCase_ = sigma_max if sigma_max is not None else self.config.sigma_max
UpperCamelCase_ = sampling_eps if sampling_eps is not None else self.config.sampling_eps
if self.timesteps is None:
self.set_timesteps(_UpperCAmelCase , _UpperCAmelCase )
UpperCamelCase_ = sigma_min * (sigma_max / sigma_min) ** (self.timesteps / sampling_eps)
UpperCamelCase_ = torch.exp(torch.linspace(math.log(_UpperCAmelCase ) , math.log(_UpperCAmelCase ) , _UpperCAmelCase ) )
UpperCamelCase_ = torch.tensor([sigma_min * (sigma_max / sigma_min) ** t for t in self.timesteps] )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> List[Any]:
return torch.where(
timesteps == 0 , torch.zeros_like(t.to(timesteps.device ) ) , self.discrete_sigmas[timesteps - 1].to(timesteps.device ) , )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase = None , _UpperCAmelCase = True , ) -> Union[SdeVeOutput, Tuple]:
if self.timesteps is None:
raise ValueError(
'`self.timesteps` is not set, you need to run \'set_timesteps\' after creating the scheduler' )
UpperCamelCase_ = timestep * torch.ones(
sample.shape[0] , device=sample.device ) # torch.repeat_interleave(timestep, sample.shape[0])
UpperCamelCase_ = (timestep * (len(self.timesteps ) - 1)).long()
# mps requires indices to be in the same device, so we use cpu as is the default with cuda
UpperCamelCase_ = timesteps.to(self.discrete_sigmas.device )
UpperCamelCase_ = self.discrete_sigmas[timesteps].to(sample.device )
UpperCamelCase_ = self.get_adjacent_sigma(_UpperCAmelCase , _UpperCAmelCase ).to(sample.device )
UpperCamelCase_ = torch.zeros_like(_UpperCAmelCase )
UpperCamelCase_ = (sigma**2 - adjacent_sigma**2) ** 0.5
# equation 6 in the paper: the model_output modeled by the network is grad_x log pt(x)
# also equation 47 shows the analog from SDE models to ancestral sampling methods
UpperCamelCase_ = diffusion.flatten()
while len(diffusion.shape ) < len(sample.shape ):
UpperCamelCase_ = diffusion.unsqueeze(-1 )
UpperCamelCase_ = drift - diffusion**2 * model_output
# equation 6: sample noise for the diffusion term of
UpperCamelCase_ = randn_tensor(
sample.shape , layout=sample.layout , generator=_UpperCAmelCase , device=sample.device , dtype=sample.dtype )
UpperCamelCase_ = sample - drift # subtract because `dt` is a small negative timestep
# TODO is the variable diffusion the correct scaling term for the noise?
UpperCamelCase_ = prev_sample_mean + diffusion * noise # add impact of diffusion field g
if not return_dict:
return (prev_sample, prev_sample_mean)
return SdeVeOutput(prev_sample=_UpperCAmelCase , prev_sample_mean=_UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase = None , _UpperCAmelCase = True , ) -> Union[SchedulerOutput, Tuple]:
if self.timesteps is None:
raise ValueError(
'`self.timesteps` is not set, you need to run \'set_timesteps\' after creating the scheduler' )
# For small batch sizes, the paper "suggest replacing norm(z) with sqrt(d), where d is the dim. of z"
# sample noise for correction
UpperCamelCase_ = randn_tensor(sample.shape , layout=sample.layout , generator=_UpperCAmelCase ).to(sample.device )
# compute step size from the model_output, the noise, and the snr
UpperCamelCase_ = torch.norm(model_output.reshape(model_output.shape[0] , -1 ) , dim=-1 ).mean()
UpperCamelCase_ = torch.norm(noise.reshape(noise.shape[0] , -1 ) , dim=-1 ).mean()
UpperCamelCase_ = (self.config.snr * noise_norm / grad_norm) ** 2 * 2
UpperCamelCase_ = step_size * torch.ones(sample.shape[0] ).to(sample.device )
# self.repeat_scalar(step_size, sample.shape[0])
# compute corrected sample: model_output term and noise term
UpperCamelCase_ = step_size.flatten()
while len(step_size.shape ) < len(sample.shape ):
UpperCamelCase_ = step_size.unsqueeze(-1 )
UpperCamelCase_ = sample + step_size * model_output
UpperCamelCase_ = prev_sample_mean + ((step_size * 2) ** 0.5) * noise
if not return_dict:
return (prev_sample,)
return SchedulerOutput(prev_sample=_UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , ) -> torch.FloatTensor:
# Make sure sigmas and timesteps have the same device and dtype as original_samples
UpperCamelCase_ = timesteps.to(original_samples.device )
UpperCamelCase_ = self.discrete_sigmas.to(original_samples.device )[timesteps]
UpperCamelCase_ = (
noise * sigmas[:, None, None, None]
if noise is not None
else torch.randn_like(_UpperCAmelCase ) * sigmas[:, None, None, None]
)
UpperCamelCase_ = noise + original_samples
return noisy_samples
def __len__( self ) -> Optional[int]:
return self.config.num_train_timesteps
| 23 | 1 |
from ...configuration_utils import PretrainedConfig
from ...utils import logging
snake_case__ : int = logging.get_logger(__name__)
snake_case__ : Any = {
"""unc-nlp/lxmert-base-uncased""": """https://huggingface.co/unc-nlp/lxmert-base-uncased/resolve/main/config.json""",
}
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
A_ = """lxmert"""
A_ = {}
def __init__( self , _UpperCAmelCase=30522 , _UpperCAmelCase=768 , _UpperCAmelCase=12 , _UpperCAmelCase=9500 , _UpperCAmelCase=1600 , _UpperCAmelCase=400 , _UpperCAmelCase=3072 , _UpperCAmelCase="gelu" , _UpperCAmelCase=0.1 , _UpperCAmelCase=0.1 , _UpperCAmelCase=512 , _UpperCAmelCase=2 , _UpperCAmelCase=0.0_2 , _UpperCAmelCase=1e-12 , _UpperCAmelCase=9 , _UpperCAmelCase=5 , _UpperCAmelCase=5 , _UpperCAmelCase=2048 , _UpperCAmelCase=4 , _UpperCAmelCase=6.6_7 , _UpperCAmelCase=True , _UpperCAmelCase=True , _UpperCAmelCase=True , _UpperCAmelCase=True , _UpperCAmelCase=True , _UpperCAmelCase=True , _UpperCAmelCase=True , **_UpperCAmelCase , ) -> str:
UpperCamelCase_ = vocab_size
UpperCamelCase_ = hidden_size
UpperCamelCase_ = num_attention_heads
UpperCamelCase_ = hidden_act
UpperCamelCase_ = intermediate_size
UpperCamelCase_ = hidden_dropout_prob
UpperCamelCase_ = attention_probs_dropout_prob
UpperCamelCase_ = max_position_embeddings
UpperCamelCase_ = type_vocab_size
UpperCamelCase_ = initializer_range
UpperCamelCase_ = layer_norm_eps
UpperCamelCase_ = num_qa_labels
UpperCamelCase_ = num_object_labels
UpperCamelCase_ = num_attr_labels
UpperCamelCase_ = l_layers
UpperCamelCase_ = x_layers
UpperCamelCase_ = r_layers
UpperCamelCase_ = visual_feat_dim
UpperCamelCase_ = visual_pos_dim
UpperCamelCase_ = visual_loss_normalizer
UpperCamelCase_ = task_matched
UpperCamelCase_ = task_mask_lm
UpperCamelCase_ = task_obj_predict
UpperCamelCase_ = task_qa
UpperCamelCase_ = visual_obj_loss
UpperCamelCase_ = visual_attr_loss
UpperCamelCase_ = visual_feat_loss
UpperCamelCase_ = {'vision': r_layers, 'cross_encoder': x_layers, 'language': l_layers}
super().__init__(**_UpperCAmelCase )
| 23 |
from typing import TYPE_CHECKING
from ...utils import OptionalDependencyNotAvailable, _LazyModule, is_torch_available
snake_case__ : Optional[int] = {
"""configuration_pegasus_x""": ["""PEGASUS_X_PRETRAINED_CONFIG_ARCHIVE_MAP""", """PegasusXConfig"""],
}
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
snake_case__ : Dict = [
"""PEGASUS_X_PRETRAINED_MODEL_ARCHIVE_LIST""",
"""PegasusXForConditionalGeneration""",
"""PegasusXModel""",
"""PegasusXPreTrainedModel""",
]
if TYPE_CHECKING:
from .configuration_pegasus_x import PEGASUS_X_PRETRAINED_CONFIG_ARCHIVE_MAP, PegasusXConfig
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_pegasus_x import (
PEGASUS_X_PRETRAINED_MODEL_ARCHIVE_LIST,
PegasusXForConditionalGeneration,
PegasusXModel,
PegasusXPreTrainedModel,
)
else:
import sys
snake_case__ : Optional[Any] = _LazyModule(__name__, globals()["""__file__"""], _import_structure, module_spec=__spec__)
| 23 | 1 |
import os
import re
import warnings
from shutil import copyfile
from typing import List, Optional, Tuple
from ...tokenization_utils_fast import PreTrainedTokenizerFast
from ...utils import is_sentencepiece_available, logging
if is_sentencepiece_available():
from .tokenization_ta import TaTokenizer
else:
snake_case__ : str = None
snake_case__ : Optional[int] = logging.get_logger(__name__)
snake_case__ : Dict = {"""vocab_file""": """spiece.model""", """tokenizer_file""": """tokenizer.json"""}
snake_case__ : Any = {
"""vocab_file""": {
"""t5-small""": """https://huggingface.co/t5-small/resolve/main/spiece.model""",
"""t5-base""": """https://huggingface.co/t5-base/resolve/main/spiece.model""",
"""t5-large""": """https://huggingface.co/t5-large/resolve/main/spiece.model""",
"""t5-3b""": """https://huggingface.co/t5-3b/resolve/main/spiece.model""",
"""t5-11b""": """https://huggingface.co/t5-11b/resolve/main/spiece.model""",
},
"""tokenizer_file""": {
"""t5-small""": """https://huggingface.co/t5-small/resolve/main/tokenizer.json""",
"""t5-base""": """https://huggingface.co/t5-base/resolve/main/tokenizer.json""",
"""t5-large""": """https://huggingface.co/t5-large/resolve/main/tokenizer.json""",
"""t5-3b""": """https://huggingface.co/t5-3b/resolve/main/tokenizer.json""",
"""t5-11b""": """https://huggingface.co/t5-11b/resolve/main/tokenizer.json""",
},
}
# TODO(PVP) - this should be removed in Transformers v5
snake_case__ : List[Any] = {
"""t5-small""": 5_1_2,
"""t5-base""": 5_1_2,
"""t5-large""": 5_1_2,
"""t5-3b""": 5_1_2,
"""t5-11b""": 5_1_2,
}
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
A_ = VOCAB_FILES_NAMES
A_ = PRETRAINED_VOCAB_FILES_MAP
A_ = PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES
A_ = ["""input_ids""", """attention_mask"""]
A_ = TaTokenizer
A_ = []
def __init__( self , _UpperCAmelCase=None , _UpperCAmelCase=None , _UpperCAmelCase="</s>" , _UpperCAmelCase="<unk>" , _UpperCAmelCase="<pad>" , _UpperCAmelCase=100 , _UpperCAmelCase=None , **_UpperCAmelCase , ) -> Dict:
# Add extra_ids to the special token list
if extra_ids > 0 and additional_special_tokens is None:
UpperCamelCase_ = [f"""<extra_id_{i}>""" for i in range(_UpperCAmelCase )]
elif extra_ids > 0 and additional_special_tokens is not None:
# Check that we have the right number of extra special tokens
UpperCamelCase_ = len(set(filter(lambda _UpperCAmelCase : bool('extra_id_' in str(_UpperCAmelCase ) ) , _UpperCAmelCase ) ) )
if extra_tokens != extra_ids:
raise ValueError(
f"""Both extra_ids ({extra_ids}) and additional_special_tokens ({additional_special_tokens}) are"""
' provided to T5Tokenizer. In this case the additional_special_tokens must include the extra_ids'
' tokens' )
super().__init__(
_UpperCAmelCase , tokenizer_file=_UpperCAmelCase , eos_token=_UpperCAmelCase , unk_token=_UpperCAmelCase , pad_token=_UpperCAmelCase , extra_ids=_UpperCAmelCase , additional_special_tokens=_UpperCAmelCase , **_UpperCAmelCase , )
UpperCamelCase_ = vocab_file
UpperCamelCase_ = False if not self.vocab_file else True
UpperCamelCase_ = extra_ids
@staticmethod
def _UpperCAmelCase ( _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase ) -> List[str]:
if pretrained_model_name_or_path in TaTokenizerFast.max_model_input_sizes:
UpperCamelCase_ = TaTokenizerFast.max_model_input_sizes[pretrained_model_name_or_path]
if init_max_model_length is not None and init_max_model_length != max_model_length:
return init_max_model_length
elif init_max_model_length is None:
warnings.warn(
'This tokenizer was incorrectly instantiated with a model max length of'
f""" {deprecated_max_model_length} which will be corrected in Transformers v5.\nFor now, this"""
' behavior is kept to avoid breaking backwards compatibility when padding/encoding with'
' `truncation is True`.\n- Be aware that you SHOULD NOT rely on'
f""" {pretrained_model_name_or_path} automatically truncating your input to"""
f""" {deprecated_max_model_length} when padding/encoding.\n- If you want to encode/pad to sequences"""
f""" longer than {deprecated_max_model_length} you can either instantiate this tokenizer with"""
' `model_max_length` or pass `max_length` when encoding/padding.\n- To avoid this warning, please'
' instantiate this tokenizer with `model_max_length` set to your preferred value.' , _UpperCAmelCase , )
return max_model_length
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase = None ) -> Tuple[str]:
if not self.can_save_slow_tokenizer:
raise ValueError(
'Your fast tokenizer does not have the necessary information to save the vocabulary for a slow '
'tokenizer.' )
if not os.path.isdir(_UpperCAmelCase ):
logger.error(f"""Vocabulary path ({save_directory}) should be a directory""" )
return
UpperCamelCase_ = os.path.join(
_UpperCAmelCase , (filename_prefix + '-' if filename_prefix else '') + VOCAB_FILES_NAMES['vocab_file'] )
if os.path.abspath(self.vocab_file ) != os.path.abspath(_UpperCAmelCase ):
copyfile(self.vocab_file , _UpperCAmelCase )
logger.info(f"""Copy vocab file to {out_vocab_file}""" )
return (out_vocab_file,)
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase = None ) -> List[int]:
UpperCamelCase_ = token_ids_a + [self.eos_token_id]
if token_ids_a is None:
return self.prefix_tokens + token_ids_a
else:
UpperCamelCase_ = token_ids_a + [self.eos_token_id]
return self.prefix_tokens + token_ids_a + token_ids_a
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase = None ) -> List[int]:
UpperCamelCase_ = [self.eos_token_id]
if token_ids_a is None:
return len(token_ids_a + eos ) * [0]
return len(token_ids_a + eos + token_ids_a + eos ) * [0]
def _UpperCAmelCase ( self ) -> List[str]:
return list(
set(filter(lambda _UpperCAmelCase : bool(re.search(R'<extra_id_\d+>' , _UpperCAmelCase ) ) is not None , self.additional_special_tokens ) ) )
def _UpperCAmelCase ( self ) -> Optional[int]:
return [self.convert_tokens_to_ids(_UpperCAmelCase ) for token in self.get_sentinel_tokens()]
| 23 |
import datasets
from .evaluate import evaluate
snake_case__ : int = """\
@article{hendrycks2021cuad,
title={CUAD: An Expert-Annotated NLP Dataset for Legal Contract Review},
author={Dan Hendrycks and Collin Burns and Anya Chen and Spencer Ball},
journal={arXiv preprint arXiv:2103.06268},
year={2021}
}
"""
snake_case__ : Union[str, Any] = """
This metric wrap the official scoring script for version 1 of the Contract
Understanding Atticus Dataset (CUAD).
Contract Understanding Atticus Dataset (CUAD) v1 is a corpus of more than 13,000 labels in 510
commercial legal contracts that have been manually labeled to identify 41 categories of important
clauses that lawyers look for when reviewing contracts in connection with corporate transactions.
"""
snake_case__ : Any = """
Computes CUAD scores (EM, F1, AUPR, Precision@80%Recall, and Precision@90%Recall).
Args:
predictions: List of question-answers dictionaries with the following key-values:
- 'id': id of the question-answer pair as given in the references (see below)
- 'prediction_text': list of possible texts for the answer, as a list of strings
depending on a threshold on the confidence probability of each prediction.
references: List of question-answers dictionaries with the following key-values:
- 'id': id of the question-answer pair (see above),
- 'answers': a Dict in the CUAD dataset format
{
'text': list of possible texts for the answer, as a list of strings
'answer_start': list of start positions for the answer, as a list of ints
}
Note that answer_start values are not taken into account to compute the metric.
Returns:
'exact_match': Exact match (the normalized answer exactly match the gold answer)
'f1': The F-score of predicted tokens versus the gold answer
'aupr': Area Under the Precision-Recall curve
'prec_at_80_recall': Precision at 80% recall
'prec_at_90_recall': Precision at 90% recall
Examples:
>>> predictions = [{'prediction_text': ['The seller:', 'The buyer/End-User: Shenzhen LOHAS Supply Chain Management Co., Ltd.'], 'id': 'LohaCompanyltd_20191209_F-1_EX-10.16_11917878_EX-10.16_Supply Agreement__Parties'}]
>>> references = [{'answers': {'answer_start': [143, 49], 'text': ['The seller:', 'The buyer/End-User: Shenzhen LOHAS Supply Chain Management Co., Ltd.']}, 'id': 'LohaCompanyltd_20191209_F-1_EX-10.16_11917878_EX-10.16_Supply Agreement__Parties'}]
>>> cuad_metric = datasets.load_metric(\"cuad\")
>>> results = cuad_metric.compute(predictions=predictions, references=references)
>>> print(results)
{'exact_match': 100.0, 'f1': 100.0, 'aupr': 0.0, 'prec_at_80_recall': 1.0, 'prec_at_90_recall': 1.0}
"""
@datasets.utils.file_utils.add_start_docstrings(_DESCRIPTION , _KWARGS_DESCRIPTION )
class _a ( datasets.Metric ):
"""simple docstring"""
def _UpperCAmelCase ( self ) -> Optional[int]:
return datasets.MetricInfo(
description=_DESCRIPTION , citation=_CITATION , inputs_description=_KWARGS_DESCRIPTION , features=datasets.Features(
{
'predictions': {
'id': datasets.Value('string' ),
'prediction_text': datasets.features.Sequence(datasets.Value('string' ) ),
},
'references': {
'id': datasets.Value('string' ),
'answers': datasets.features.Sequence(
{
'text': datasets.Value('string' ),
'answer_start': datasets.Value('int32' ),
} ),
},
} ) , codebase_urls=['https://www.atticusprojectai.org/cuad'] , reference_urls=['https://www.atticusprojectai.org/cuad'] , )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> Dict:
UpperCamelCase_ = {prediction['id']: prediction['prediction_text'] for prediction in predictions}
UpperCamelCase_ = [
{
'paragraphs': [
{
'qas': [
{
'answers': [{'text': answer_text} for answer_text in ref['answers']['text']],
'id': ref['id'],
}
for ref in references
]
}
]
}
]
UpperCamelCase_ = evaluate(dataset=_UpperCAmelCase , predictions=_UpperCAmelCase )
return score
| 23 | 1 |
import json
from typing import List, Optional, Tuple
from tokenizers import normalizers
from ...tokenization_utils_fast import PreTrainedTokenizerFast
from ...utils import logging
from .tokenization_bert import BertTokenizer
snake_case__ : Optional[int] = logging.get_logger(__name__)
snake_case__ : Optional[Any] = {"""vocab_file""": """vocab.txt""", """tokenizer_file""": """tokenizer.json"""}
snake_case__ : Any = {
"""vocab_file""": {
"""bert-base-uncased""": """https://huggingface.co/bert-base-uncased/resolve/main/vocab.txt""",
"""bert-large-uncased""": """https://huggingface.co/bert-large-uncased/resolve/main/vocab.txt""",
"""bert-base-cased""": """https://huggingface.co/bert-base-cased/resolve/main/vocab.txt""",
"""bert-large-cased""": """https://huggingface.co/bert-large-cased/resolve/main/vocab.txt""",
"""bert-base-multilingual-uncased""": (
"""https://huggingface.co/bert-base-multilingual-uncased/resolve/main/vocab.txt"""
),
"""bert-base-multilingual-cased""": """https://huggingface.co/bert-base-multilingual-cased/resolve/main/vocab.txt""",
"""bert-base-chinese""": """https://huggingface.co/bert-base-chinese/resolve/main/vocab.txt""",
"""bert-base-german-cased""": """https://huggingface.co/bert-base-german-cased/resolve/main/vocab.txt""",
"""bert-large-uncased-whole-word-masking""": (
"""https://huggingface.co/bert-large-uncased-whole-word-masking/resolve/main/vocab.txt"""
),
"""bert-large-cased-whole-word-masking""": (
"""https://huggingface.co/bert-large-cased-whole-word-masking/resolve/main/vocab.txt"""
),
"""bert-large-uncased-whole-word-masking-finetuned-squad""": (
"""https://huggingface.co/bert-large-uncased-whole-word-masking-finetuned-squad/resolve/main/vocab.txt"""
),
"""bert-large-cased-whole-word-masking-finetuned-squad""": (
"""https://huggingface.co/bert-large-cased-whole-word-masking-finetuned-squad/resolve/main/vocab.txt"""
),
"""bert-base-cased-finetuned-mrpc""": (
"""https://huggingface.co/bert-base-cased-finetuned-mrpc/resolve/main/vocab.txt"""
),
"""bert-base-german-dbmdz-cased""": """https://huggingface.co/bert-base-german-dbmdz-cased/resolve/main/vocab.txt""",
"""bert-base-german-dbmdz-uncased""": (
"""https://huggingface.co/bert-base-german-dbmdz-uncased/resolve/main/vocab.txt"""
),
"""TurkuNLP/bert-base-finnish-cased-v1""": (
"""https://huggingface.co/TurkuNLP/bert-base-finnish-cased-v1/resolve/main/vocab.txt"""
),
"""TurkuNLP/bert-base-finnish-uncased-v1""": (
"""https://huggingface.co/TurkuNLP/bert-base-finnish-uncased-v1/resolve/main/vocab.txt"""
),
"""wietsedv/bert-base-dutch-cased""": (
"""https://huggingface.co/wietsedv/bert-base-dutch-cased/resolve/main/vocab.txt"""
),
},
"""tokenizer_file""": {
"""bert-base-uncased""": """https://huggingface.co/bert-base-uncased/resolve/main/tokenizer.json""",
"""bert-large-uncased""": """https://huggingface.co/bert-large-uncased/resolve/main/tokenizer.json""",
"""bert-base-cased""": """https://huggingface.co/bert-base-cased/resolve/main/tokenizer.json""",
"""bert-large-cased""": """https://huggingface.co/bert-large-cased/resolve/main/tokenizer.json""",
"""bert-base-multilingual-uncased""": (
"""https://huggingface.co/bert-base-multilingual-uncased/resolve/main/tokenizer.json"""
),
"""bert-base-multilingual-cased""": (
"""https://huggingface.co/bert-base-multilingual-cased/resolve/main/tokenizer.json"""
),
"""bert-base-chinese""": """https://huggingface.co/bert-base-chinese/resolve/main/tokenizer.json""",
"""bert-base-german-cased""": """https://huggingface.co/bert-base-german-cased/resolve/main/tokenizer.json""",
"""bert-large-uncased-whole-word-masking""": (
"""https://huggingface.co/bert-large-uncased-whole-word-masking/resolve/main/tokenizer.json"""
),
"""bert-large-cased-whole-word-masking""": (
"""https://huggingface.co/bert-large-cased-whole-word-masking/resolve/main/tokenizer.json"""
),
"""bert-large-uncased-whole-word-masking-finetuned-squad""": (
"""https://huggingface.co/bert-large-uncased-whole-word-masking-finetuned-squad/resolve/main/tokenizer.json"""
),
"""bert-large-cased-whole-word-masking-finetuned-squad""": (
"""https://huggingface.co/bert-large-cased-whole-word-masking-finetuned-squad/resolve/main/tokenizer.json"""
),
"""bert-base-cased-finetuned-mrpc""": (
"""https://huggingface.co/bert-base-cased-finetuned-mrpc/resolve/main/tokenizer.json"""
),
"""bert-base-german-dbmdz-cased""": (
"""https://huggingface.co/bert-base-german-dbmdz-cased/resolve/main/tokenizer.json"""
),
"""bert-base-german-dbmdz-uncased""": (
"""https://huggingface.co/bert-base-german-dbmdz-uncased/resolve/main/tokenizer.json"""
),
"""TurkuNLP/bert-base-finnish-cased-v1""": (
"""https://huggingface.co/TurkuNLP/bert-base-finnish-cased-v1/resolve/main/tokenizer.json"""
),
"""TurkuNLP/bert-base-finnish-uncased-v1""": (
"""https://huggingface.co/TurkuNLP/bert-base-finnish-uncased-v1/resolve/main/tokenizer.json"""
),
"""wietsedv/bert-base-dutch-cased""": (
"""https://huggingface.co/wietsedv/bert-base-dutch-cased/resolve/main/tokenizer.json"""
),
},
}
snake_case__ : List[str] = {
"""bert-base-uncased""": 5_1_2,
"""bert-large-uncased""": 5_1_2,
"""bert-base-cased""": 5_1_2,
"""bert-large-cased""": 5_1_2,
"""bert-base-multilingual-uncased""": 5_1_2,
"""bert-base-multilingual-cased""": 5_1_2,
"""bert-base-chinese""": 5_1_2,
"""bert-base-german-cased""": 5_1_2,
"""bert-large-uncased-whole-word-masking""": 5_1_2,
"""bert-large-cased-whole-word-masking""": 5_1_2,
"""bert-large-uncased-whole-word-masking-finetuned-squad""": 5_1_2,
"""bert-large-cased-whole-word-masking-finetuned-squad""": 5_1_2,
"""bert-base-cased-finetuned-mrpc""": 5_1_2,
"""bert-base-german-dbmdz-cased""": 5_1_2,
"""bert-base-german-dbmdz-uncased""": 5_1_2,
"""TurkuNLP/bert-base-finnish-cased-v1""": 5_1_2,
"""TurkuNLP/bert-base-finnish-uncased-v1""": 5_1_2,
"""wietsedv/bert-base-dutch-cased""": 5_1_2,
}
snake_case__ : Tuple = {
"""bert-base-uncased""": {"""do_lower_case""": True},
"""bert-large-uncased""": {"""do_lower_case""": True},
"""bert-base-cased""": {"""do_lower_case""": False},
"""bert-large-cased""": {"""do_lower_case""": False},
"""bert-base-multilingual-uncased""": {"""do_lower_case""": True},
"""bert-base-multilingual-cased""": {"""do_lower_case""": False},
"""bert-base-chinese""": {"""do_lower_case""": False},
"""bert-base-german-cased""": {"""do_lower_case""": False},
"""bert-large-uncased-whole-word-masking""": {"""do_lower_case""": True},
"""bert-large-cased-whole-word-masking""": {"""do_lower_case""": False},
"""bert-large-uncased-whole-word-masking-finetuned-squad""": {"""do_lower_case""": True},
"""bert-large-cased-whole-word-masking-finetuned-squad""": {"""do_lower_case""": False},
"""bert-base-cased-finetuned-mrpc""": {"""do_lower_case""": False},
"""bert-base-german-dbmdz-cased""": {"""do_lower_case""": False},
"""bert-base-german-dbmdz-uncased""": {"""do_lower_case""": True},
"""TurkuNLP/bert-base-finnish-cased-v1""": {"""do_lower_case""": False},
"""TurkuNLP/bert-base-finnish-uncased-v1""": {"""do_lower_case""": True},
"""wietsedv/bert-base-dutch-cased""": {"""do_lower_case""": False},
}
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
A_ = VOCAB_FILES_NAMES
A_ = PRETRAINED_VOCAB_FILES_MAP
A_ = PRETRAINED_INIT_CONFIGURATION
A_ = PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES
A_ = BertTokenizer
def __init__( self , _UpperCAmelCase=None , _UpperCAmelCase=None , _UpperCAmelCase=True , _UpperCAmelCase="[UNK]" , _UpperCAmelCase="[SEP]" , _UpperCAmelCase="[PAD]" , _UpperCAmelCase="[CLS]" , _UpperCAmelCase="[MASK]" , _UpperCAmelCase=True , _UpperCAmelCase=None , **_UpperCAmelCase , ) -> List[Any]:
super().__init__(
_UpperCAmelCase , tokenizer_file=_UpperCAmelCase , do_lower_case=_UpperCAmelCase , unk_token=_UpperCAmelCase , sep_token=_UpperCAmelCase , pad_token=_UpperCAmelCase , cls_token=_UpperCAmelCase , mask_token=_UpperCAmelCase , tokenize_chinese_chars=_UpperCAmelCase , strip_accents=_UpperCAmelCase , **_UpperCAmelCase , )
UpperCamelCase_ = json.loads(self.backend_tokenizer.normalizer.__getstate__() )
if (
normalizer_state.get('lowercase' , _UpperCAmelCase ) != do_lower_case
or normalizer_state.get('strip_accents' , _UpperCAmelCase ) != strip_accents
or normalizer_state.get('handle_chinese_chars' , _UpperCAmelCase ) != tokenize_chinese_chars
):
UpperCamelCase_ = getattr(_UpperCAmelCase , normalizer_state.pop('type' ) )
UpperCamelCase_ = do_lower_case
UpperCamelCase_ = strip_accents
UpperCamelCase_ = tokenize_chinese_chars
UpperCamelCase_ = normalizer_class(**_UpperCAmelCase )
UpperCamelCase_ = do_lower_case
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase=None ) -> Union[str, Any]:
UpperCamelCase_ = [self.cls_token_id] + token_ids_a + [self.sep_token_id]
if token_ids_a:
output += token_ids_a + [self.sep_token_id]
return output
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase = None ) -> List[int]:
UpperCamelCase_ = [self.sep_token_id]
UpperCamelCase_ = [self.cls_token_id]
if token_ids_a is None:
return len(cls + token_ids_a + sep ) * [0]
return len(cls + token_ids_a + sep ) * [0] + len(token_ids_a + sep ) * [1]
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase = None ) -> Tuple[str]:
UpperCamelCase_ = self._tokenizer.model.save(_UpperCAmelCase , name=_UpperCAmelCase )
return tuple(_UpperCAmelCase )
| 23 |
import os
import tempfile
from functools import partial
from unittest import TestCase
from unittest.mock import patch
import datasets
import datasets.config
from .utils import require_beam
class _a ( datasets.BeamBasedBuilder ):
"""simple docstring"""
def _UpperCAmelCase ( self ) -> List[str]:
return datasets.DatasetInfo(
features=datasets.Features({'content': datasets.Value('string' )} ) , supervised_keys=_UpperCAmelCase , )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> Optional[Any]:
return [datasets.SplitGenerator(name=datasets.Split.TRAIN , gen_kwargs={'examples': get_test_dummy_examples()} )]
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> Tuple:
import apache_beam as beam
return pipeline | "Load Examples" >> beam.Create(_UpperCAmelCase )
class _a ( datasets.BeamBasedBuilder ):
"""simple docstring"""
def _UpperCAmelCase ( self ) -> Any:
return datasets.DatasetInfo(
features=datasets.Features({'a': datasets.Sequence({'b': datasets.Value('string' )} )} ) , supervised_keys=_UpperCAmelCase , )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> Tuple:
return [
datasets.SplitGenerator(name=datasets.Split.TRAIN , gen_kwargs={'examples': get_test_nested_examples()} )
]
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> Union[str, Any]:
import apache_beam as beam
return pipeline | "Load Examples" >> beam.Create(_UpperCAmelCase )
def _snake_case ():
return [(i, {"content": content}) for i, content in enumerate(['foo', 'bar', 'foobar'])]
def _snake_case ():
return [(i, {"a": {"b": [content]}}) for i, content in enumerate(['foo', 'bar', 'foobar'])]
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
@require_beam
def _UpperCAmelCase ( self ) -> Dict:
UpperCamelCase_ = len(get_test_dummy_examples() )
with tempfile.TemporaryDirectory() as tmp_cache_dir:
UpperCamelCase_ = DummyBeamDataset(cache_dir=_UpperCAmelCase , beam_runner='DirectRunner' )
builder.download_and_prepare()
self.assertTrue(
os.path.exists(
os.path.join(_UpperCAmelCase , builder.name , 'default' , '0.0.0' , f"""{builder.name}-train.arrow""" ) ) )
self.assertDictEqual(builder.info.features , datasets.Features({'content': datasets.Value('string' )} ) )
UpperCamelCase_ = builder.as_dataset()
self.assertEqual(dset['train'].num_rows , _UpperCAmelCase )
self.assertEqual(dset['train'].info.splits['train'].num_examples , _UpperCAmelCase )
self.assertDictEqual(dset['train'][0] , get_test_dummy_examples()[0][1] )
self.assertDictEqual(
dset['train'][expected_num_examples - 1] , get_test_dummy_examples()[expected_num_examples - 1][1] )
self.assertTrue(
os.path.exists(os.path.join(_UpperCAmelCase , builder.name , 'default' , '0.0.0' , 'dataset_info.json' ) ) )
del dset
@require_beam
def _UpperCAmelCase ( self ) -> List[str]:
import apache_beam as beam
UpperCamelCase_ = beam.io.parquetio.WriteToParquet
UpperCamelCase_ = len(get_test_dummy_examples() )
with tempfile.TemporaryDirectory() as tmp_cache_dir:
UpperCamelCase_ = DummyBeamDataset(cache_dir=_UpperCAmelCase , beam_runner='DirectRunner' )
with patch('apache_beam.io.parquetio.WriteToParquet' ) as write_parquet_mock:
UpperCamelCase_ = partial(_UpperCAmelCase , num_shards=2 )
builder.download_and_prepare()
self.assertTrue(
os.path.exists(
os.path.join(
_UpperCAmelCase , builder.name , 'default' , '0.0.0' , f"""{builder.name}-train-00000-of-00002.arrow""" ) ) )
self.assertTrue(
os.path.exists(
os.path.join(
_UpperCAmelCase , builder.name , 'default' , '0.0.0' , f"""{builder.name}-train-00000-of-00002.arrow""" ) ) )
self.assertDictEqual(builder.info.features , datasets.Features({'content': datasets.Value('string' )} ) )
UpperCamelCase_ = builder.as_dataset()
self.assertEqual(dset['train'].num_rows , _UpperCAmelCase )
self.assertEqual(dset['train'].info.splits['train'].num_examples , _UpperCAmelCase )
# Order is not preserved when sharding, so we just check that all the elements are there
self.assertListEqual(sorted(dset['train']['content'] ) , sorted(['foo', 'bar', 'foobar'] ) )
self.assertTrue(
os.path.exists(os.path.join(_UpperCAmelCase , builder.name , 'default' , '0.0.0' , 'dataset_info.json' ) ) )
del dset
@require_beam
def _UpperCAmelCase ( self ) -> Any:
with tempfile.TemporaryDirectory() as tmp_cache_dir:
UpperCamelCase_ = DummyBeamDataset(cache_dir=_UpperCAmelCase )
self.assertRaises(datasets.builder.MissingBeamOptions , builder.download_and_prepare )
@require_beam
def _UpperCAmelCase ( self ) -> Union[str, Any]:
UpperCamelCase_ = len(get_test_nested_examples() )
with tempfile.TemporaryDirectory() as tmp_cache_dir:
UpperCamelCase_ = NestedBeamDataset(cache_dir=_UpperCAmelCase , beam_runner='DirectRunner' )
builder.download_and_prepare()
self.assertTrue(
os.path.exists(
os.path.join(_UpperCAmelCase , builder.name , 'default' , '0.0.0' , f"""{builder.name}-train.arrow""" ) ) )
self.assertDictEqual(
builder.info.features , datasets.Features({'a': datasets.Sequence({'b': datasets.Value('string' )} )} ) )
UpperCamelCase_ = builder.as_dataset()
self.assertEqual(dset['train'].num_rows , _UpperCAmelCase )
self.assertEqual(dset['train'].info.splits['train'].num_examples , _UpperCAmelCase )
self.assertDictEqual(dset['train'][0] , get_test_nested_examples()[0][1] )
self.assertDictEqual(
dset['train'][expected_num_examples - 1] , get_test_nested_examples()[expected_num_examples - 1][1] )
self.assertTrue(
os.path.exists(os.path.join(_UpperCAmelCase , builder.name , 'default' , '0.0.0' , 'dataset_info.json' ) ) )
del dset
| 23 | 1 |
from transformers import HfArgumentParser, TensorFlowBenchmark, TensorFlowBenchmarkArguments
def _snake_case ():
UpperCamelCase_ = HfArgumentParser(__lowercase)
UpperCamelCase_ = parser.parse_args_into_dataclasses()[0]
UpperCamelCase_ = TensorFlowBenchmark(args=__lowercase)
try:
UpperCamelCase_ = parser.parse_args_into_dataclasses()[0]
except ValueError as e:
UpperCamelCase_ = 'Arg --no_{0} is no longer used, please use --no-{0} instead.'
UpperCamelCase_ = ' '.join(str(__lowercase).split(' ')[:-1])
UpperCamelCase_ = ''
UpperCamelCase_ = eval(str(__lowercase).split(' ')[-1])
UpperCamelCase_ = []
for arg in depreciated_args:
# arg[2:] removes '--'
if arg[2:] in TensorFlowBenchmark.deprecated_args:
# arg[5:] removes '--no_'
full_error_msg += arg_error_msg.format(arg[5:])
else:
wrong_args.append(__lowercase)
if len(__lowercase) > 0:
UpperCamelCase_ = full_error_msg + begin_error_msg + str(__lowercase)
raise ValueError(__lowercase)
benchmark.run()
if __name__ == "__main__":
main()
| 23 |
import argparse
import torch
from ...utils import logging
from . import AlbertConfig, AlbertForPreTraining, load_tf_weights_in_albert
logging.set_verbosity_info()
def _snake_case (__lowercase , __lowercase , __lowercase):
# Initialise PyTorch model
UpperCamelCase_ = AlbertConfig.from_json_file(__lowercase)
print(f"""Building PyTorch model from configuration: {config}""")
UpperCamelCase_ = AlbertForPreTraining(__lowercase)
# Load weights from tf checkpoint
load_tf_weights_in_albert(__lowercase , __lowercase , __lowercase)
# Save pytorch-model
print(f"""Save PyTorch model to {pytorch_dump_path}""")
torch.save(model.state_dict() , __lowercase)
if __name__ == "__main__":
snake_case__ : Tuple = argparse.ArgumentParser()
# Required parameters
parser.add_argument(
"""--tf_checkpoint_path""", default=None, type=str, required=True, help="""Path to the TensorFlow checkpoint path."""
)
parser.add_argument(
"""--albert_config_file""",
default=None,
type=str,
required=True,
help=(
"""The config json file corresponding to the pre-trained ALBERT model. \n"""
"""This specifies the model architecture."""
),
)
parser.add_argument(
"""--pytorch_dump_path""", default=None, type=str, required=True, help="""Path to the output PyTorch model."""
)
snake_case__ : str = parser.parse_args()
convert_tf_checkpoint_to_pytorch(args.tf_checkpoint_path, args.albert_config_file, args.pytorch_dump_path)
| 23 | 1 |
from typing import Optional, Union
import torch
from torch import nn
from ...configuration_utils import ConfigMixin, register_to_config
from ...models.modeling_utils import ModelMixin
class _a ( UpperCAmelCase__ , UpperCAmelCase__ ):
"""simple docstring"""
@register_to_config
def __init__( self , _UpperCAmelCase = 768 , ) -> Dict:
super().__init__()
UpperCamelCase_ = nn.Parameter(torch.zeros(1 , _UpperCAmelCase ) )
UpperCamelCase_ = nn.Parameter(torch.ones(1 , _UpperCAmelCase ) )
def _UpperCAmelCase ( self , _UpperCAmelCase = None , _UpperCAmelCase = None , ) -> List[Any]:
UpperCamelCase_ = nn.Parameter(self.mean.to(_UpperCAmelCase ).to(_UpperCAmelCase ) )
UpperCamelCase_ = nn.Parameter(self.std.to(_UpperCAmelCase ).to(_UpperCAmelCase ) )
return self
def _UpperCAmelCase ( self , _UpperCAmelCase ) -> Dict:
UpperCamelCase_ = (embeds - self.mean) * 1.0 / self.std
return embeds
def _UpperCAmelCase ( self , _UpperCAmelCase ) -> List[str]:
UpperCamelCase_ = (embeds * self.std) + self.mean
return embeds
| 23 |
from transformers import BertTokenizer, EncoderDecoderModel, SeqaSeqTrainer, SeqaSeqTrainingArguments
from transformers.testing_utils import TestCasePlus, require_torch, slow
from transformers.utils import is_datasets_available
if is_datasets_available():
import datasets
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
@slow
@require_torch
def _UpperCAmelCase ( self ) -> List[Any]:
UpperCamelCase_ = EncoderDecoderModel.from_encoder_decoder_pretrained('prajjwal1/bert-tiny' , 'prajjwal1/bert-tiny' )
UpperCamelCase_ = BertTokenizer.from_pretrained('bert-base-uncased' )
UpperCamelCase_ = bertabert.config.encoder.vocab_size
UpperCamelCase_ = tokenizer.sep_token_id
UpperCamelCase_ = tokenizer.cls_token_id
UpperCamelCase_ = 128
UpperCamelCase_ = datasets.load_dataset('cnn_dailymail' , '3.0.0' , split='train[:1%]' )
UpperCamelCase_ = datasets.load_dataset('cnn_dailymail' , '3.0.0' , split='validation[:1%]' )
UpperCamelCase_ = train_dataset.select(range(32 ) )
UpperCamelCase_ = val_dataset.select(range(16 ) )
UpperCamelCase_ = 4
def _map_to_encoder_decoder_inputs(_UpperCAmelCase ):
# Tokenizer will automatically set [BOS] <text> [EOS]
UpperCamelCase_ = tokenizer(batch['article'] , padding='max_length' , truncation=_UpperCAmelCase , max_length=512 )
UpperCamelCase_ = tokenizer(batch['highlights'] , padding='max_length' , truncation=_UpperCAmelCase , max_length=128 )
UpperCamelCase_ = inputs.input_ids
UpperCamelCase_ = inputs.attention_mask
UpperCamelCase_ = outputs.input_ids
UpperCamelCase_ = outputs.input_ids.copy()
UpperCamelCase_ = [
[-100 if token == tokenizer.pad_token_id else token for token in labels] for labels in batch['labels']
]
UpperCamelCase_ = outputs.attention_mask
assert all(len(_UpperCAmelCase ) == 512 for x in inputs.input_ids )
assert all(len(_UpperCAmelCase ) == 128 for x in outputs.input_ids )
return batch
def _compute_metrics(_UpperCAmelCase ):
UpperCamelCase_ = pred.label_ids
UpperCamelCase_ = pred.predictions
# all unnecessary tokens are removed
UpperCamelCase_ = tokenizer.batch_decode(_UpperCAmelCase , skip_special_tokens=_UpperCAmelCase )
UpperCamelCase_ = tokenizer.batch_decode(_UpperCAmelCase , skip_special_tokens=_UpperCAmelCase )
UpperCamelCase_ = sum([int(pred_str[i] == label_str[i] ) for i in range(len(_UpperCAmelCase ) )] ) / len(_UpperCAmelCase )
return {"accuracy": accuracy}
# map train dataset
UpperCamelCase_ = train_dataset.map(
_map_to_encoder_decoder_inputs , batched=_UpperCAmelCase , batch_size=_UpperCAmelCase , remove_columns=['article', 'highlights'] , )
train_dataset.set_format(
type='torch' , columns=['input_ids', 'attention_mask', 'decoder_input_ids', 'decoder_attention_mask', 'labels'] , )
# same for validation dataset
UpperCamelCase_ = val_dataset.map(
_map_to_encoder_decoder_inputs , batched=_UpperCAmelCase , batch_size=_UpperCAmelCase , remove_columns=['article', 'highlights'] , )
val_dataset.set_format(
type='torch' , columns=['input_ids', 'attention_mask', 'decoder_input_ids', 'decoder_attention_mask', 'labels'] , )
UpperCamelCase_ = self.get_auto_remove_tmp_dir()
UpperCamelCase_ = SeqaSeqTrainingArguments(
output_dir=_UpperCAmelCase , per_device_train_batch_size=_UpperCAmelCase , per_device_eval_batch_size=_UpperCAmelCase , predict_with_generate=_UpperCAmelCase , evaluation_strategy='steps' , do_train=_UpperCAmelCase , do_eval=_UpperCAmelCase , warmup_steps=0 , eval_steps=2 , logging_steps=2 , )
# instantiate trainer
UpperCamelCase_ = SeqaSeqTrainer(
model=_UpperCAmelCase , args=_UpperCAmelCase , compute_metrics=_compute_metrics , train_dataset=_UpperCAmelCase , eval_dataset=_UpperCAmelCase , tokenizer=_UpperCAmelCase , )
# start training
trainer.train()
| 23 | 1 |
import json
import pathlib
import unittest
import numpy as np
from transformers.testing_utils import require_torch, require_vision, slow
from transformers.utils import is_torch_available, is_vision_available
from ...test_image_processing_common import ImageProcessingSavingTestMixin, prepare_image_inputs
if is_torch_available():
import torch
if is_vision_available():
from PIL import Image
from transformers import ConditionalDetrImageProcessor
class _a ( unittest.TestCase ):
"""simple docstring"""
def __init__( self , _UpperCAmelCase , _UpperCAmelCase=7 , _UpperCAmelCase=3 , _UpperCAmelCase=30 , _UpperCAmelCase=400 , _UpperCAmelCase=True , _UpperCAmelCase=None , _UpperCAmelCase=True , _UpperCAmelCase=[0.5, 0.5, 0.5] , _UpperCAmelCase=[0.5, 0.5, 0.5] , _UpperCAmelCase=True , _UpperCAmelCase=1 / 255 , _UpperCAmelCase=True , ) -> str:
# by setting size["longest_edge"] > max_resolution we're effectively not testing this :p
UpperCamelCase_ = size if size is not None else {'shortest_edge': 18, 'longest_edge': 1333}
UpperCamelCase_ = parent
UpperCamelCase_ = batch_size
UpperCamelCase_ = num_channels
UpperCamelCase_ = min_resolution
UpperCamelCase_ = max_resolution
UpperCamelCase_ = do_resize
UpperCamelCase_ = size
UpperCamelCase_ = do_normalize
UpperCamelCase_ = image_mean
UpperCamelCase_ = image_std
UpperCamelCase_ = do_rescale
UpperCamelCase_ = rescale_factor
UpperCamelCase_ = do_pad
def _UpperCAmelCase ( self ) -> int:
return {
"do_resize": self.do_resize,
"size": self.size,
"do_normalize": self.do_normalize,
"image_mean": self.image_mean,
"image_std": self.image_std,
"do_rescale": self.do_rescale,
"rescale_factor": self.rescale_factor,
"do_pad": self.do_pad,
}
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase=False ) -> Tuple:
if not batched:
UpperCamelCase_ = image_inputs[0]
if isinstance(_UpperCAmelCase , Image.Image ):
UpperCamelCase_ , UpperCamelCase_ = image.size
else:
UpperCamelCase_ , UpperCamelCase_ = image.shape[1], image.shape[2]
if w < h:
UpperCamelCase_ = int(self.size['shortest_edge'] * h / w )
UpperCamelCase_ = self.size['shortest_edge']
elif w > h:
UpperCamelCase_ = self.size['shortest_edge']
UpperCamelCase_ = int(self.size['shortest_edge'] * w / h )
else:
UpperCamelCase_ = self.size['shortest_edge']
UpperCamelCase_ = self.size['shortest_edge']
else:
UpperCamelCase_ = []
for image in image_inputs:
UpperCamelCase_ , UpperCamelCase_ = self.get_expected_values([image] )
expected_values.append((expected_height, expected_width) )
UpperCamelCase_ = max(_UpperCAmelCase , key=lambda _UpperCAmelCase : item[0] )[0]
UpperCamelCase_ = max(_UpperCAmelCase , key=lambda _UpperCAmelCase : item[1] )[1]
return expected_height, expected_width
@require_torch
@require_vision
class _a ( UpperCAmelCase__ , unittest.TestCase ):
"""simple docstring"""
A_ = ConditionalDetrImageProcessor if is_vision_available() else None
def _UpperCAmelCase ( self ) -> List[str]:
UpperCamelCase_ = ConditionalDetrImageProcessingTester(self )
@property
def _UpperCAmelCase ( self ) -> int:
return self.image_processor_tester.prepare_image_processor_dict()
def _UpperCAmelCase ( self ) -> List[str]:
UpperCamelCase_ = self.image_processing_class(**self.image_processor_dict )
self.assertTrue(hasattr(_UpperCAmelCase , 'image_mean' ) )
self.assertTrue(hasattr(_UpperCAmelCase , 'image_std' ) )
self.assertTrue(hasattr(_UpperCAmelCase , 'do_normalize' ) )
self.assertTrue(hasattr(_UpperCAmelCase , 'do_resize' ) )
self.assertTrue(hasattr(_UpperCAmelCase , 'size' ) )
def _UpperCAmelCase ( self ) -> Optional[Any]:
UpperCamelCase_ = self.image_processing_class.from_dict(self.image_processor_dict )
self.assertEqual(image_processor.size , {'shortest_edge': 18, 'longest_edge': 1333} )
self.assertEqual(image_processor.do_pad , _UpperCAmelCase )
UpperCamelCase_ = self.image_processing_class.from_dict(
self.image_processor_dict , size=42 , max_size=84 , pad_and_return_pixel_mask=_UpperCAmelCase )
self.assertEqual(image_processor.size , {'shortest_edge': 42, 'longest_edge': 84} )
self.assertEqual(image_processor.do_pad , _UpperCAmelCase )
def _UpperCAmelCase ( self ) -> Dict:
pass
def _UpperCAmelCase ( self ) -> List[Any]:
# Initialize image_processing
UpperCamelCase_ = self.image_processing_class(**self.image_processor_dict )
# create random PIL images
UpperCamelCase_ = prepare_image_inputs(self.image_processor_tester , equal_resolution=_UpperCAmelCase )
for image in image_inputs:
self.assertIsInstance(_UpperCAmelCase , Image.Image )
# Test not batched input
UpperCamelCase_ = image_processing(image_inputs[0] , return_tensors='pt' ).pixel_values
UpperCamelCase_ , UpperCamelCase_ = self.image_processor_tester.get_expected_values(_UpperCAmelCase )
self.assertEqual(
encoded_images.shape , (1, self.image_processor_tester.num_channels, expected_height, expected_width) , )
# Test batched
UpperCamelCase_ , UpperCamelCase_ = self.image_processor_tester.get_expected_values(_UpperCAmelCase , batched=_UpperCAmelCase )
UpperCamelCase_ = image_processing(_UpperCAmelCase , return_tensors='pt' ).pixel_values
self.assertEqual(
encoded_images.shape , (
self.image_processor_tester.batch_size,
self.image_processor_tester.num_channels,
expected_height,
expected_width,
) , )
def _UpperCAmelCase ( self ) -> int:
# Initialize image_processing
UpperCamelCase_ = self.image_processing_class(**self.image_processor_dict )
# create random numpy tensors
UpperCamelCase_ = prepare_image_inputs(self.image_processor_tester , equal_resolution=_UpperCAmelCase , numpify=_UpperCAmelCase )
for image in image_inputs:
self.assertIsInstance(_UpperCAmelCase , np.ndarray )
# Test not batched input
UpperCamelCase_ = image_processing(image_inputs[0] , return_tensors='pt' ).pixel_values
UpperCamelCase_ , UpperCamelCase_ = self.image_processor_tester.get_expected_values(_UpperCAmelCase )
self.assertEqual(
encoded_images.shape , (1, self.image_processor_tester.num_channels, expected_height, expected_width) , )
# Test batched
UpperCamelCase_ = image_processing(_UpperCAmelCase , return_tensors='pt' ).pixel_values
UpperCamelCase_ , UpperCamelCase_ = self.image_processor_tester.get_expected_values(_UpperCAmelCase , batched=_UpperCAmelCase )
self.assertEqual(
encoded_images.shape , (
self.image_processor_tester.batch_size,
self.image_processor_tester.num_channels,
expected_height,
expected_width,
) , )
def _UpperCAmelCase ( self ) -> Any:
# Initialize image_processing
UpperCamelCase_ = self.image_processing_class(**self.image_processor_dict )
# create random PyTorch tensors
UpperCamelCase_ = prepare_image_inputs(self.image_processor_tester , equal_resolution=_UpperCAmelCase , torchify=_UpperCAmelCase )
for image in image_inputs:
self.assertIsInstance(_UpperCAmelCase , torch.Tensor )
# Test not batched input
UpperCamelCase_ = image_processing(image_inputs[0] , return_tensors='pt' ).pixel_values
UpperCamelCase_ , UpperCamelCase_ = self.image_processor_tester.get_expected_values(_UpperCAmelCase )
self.assertEqual(
encoded_images.shape , (1, self.image_processor_tester.num_channels, expected_height, expected_width) , )
# Test batched
UpperCamelCase_ = image_processing(_UpperCAmelCase , return_tensors='pt' ).pixel_values
UpperCamelCase_ , UpperCamelCase_ = self.image_processor_tester.get_expected_values(_UpperCAmelCase , batched=_UpperCAmelCase )
self.assertEqual(
encoded_images.shape , (
self.image_processor_tester.batch_size,
self.image_processor_tester.num_channels,
expected_height,
expected_width,
) , )
@slow
def _UpperCAmelCase ( self ) -> Any:
# prepare image and target
UpperCamelCase_ = Image.open('./tests/fixtures/tests_samples/COCO/000000039769.png' )
with open('./tests/fixtures/tests_samples/COCO/coco_annotations.txt' , 'r' ) as f:
UpperCamelCase_ = json.loads(f.read() )
UpperCamelCase_ = {'image_id': 39769, 'annotations': target}
# encode them
UpperCamelCase_ = ConditionalDetrImageProcessor.from_pretrained('microsoft/conditional-detr-resnet-50' )
UpperCamelCase_ = image_processing(images=_UpperCAmelCase , annotations=_UpperCAmelCase , return_tensors='pt' )
# verify pixel values
UpperCamelCase_ = torch.Size([1, 3, 800, 1066] )
self.assertEqual(encoding['pixel_values'].shape , _UpperCAmelCase )
UpperCamelCase_ = torch.tensor([0.2_7_9_6, 0.3_1_3_8, 0.3_4_8_1] )
self.assertTrue(torch.allclose(encoding['pixel_values'][0, 0, 0, :3] , _UpperCAmelCase , atol=1e-4 ) )
# verify area
UpperCamelCase_ = torch.tensor([5_8_8_7.9_6_0_0, 1_1_2_5_0.2_0_6_1, 4_8_9_3_5_3.8_4_3_8, 8_3_7_1_2_2.7_5_0_0, 1_4_7_9_6_7.5_1_5_6, 1_6_5_7_3_2.3_4_3_8] )
self.assertTrue(torch.allclose(encoding['labels'][0]['area'] , _UpperCAmelCase ) )
# verify boxes
UpperCamelCase_ = torch.Size([6, 4] )
self.assertEqual(encoding['labels'][0]['boxes'].shape , _UpperCAmelCase )
UpperCamelCase_ = torch.tensor([0.5_5_0_3, 0.2_7_6_5, 0.0_6_0_4, 0.2_2_1_5] )
self.assertTrue(torch.allclose(encoding['labels'][0]['boxes'][0] , _UpperCAmelCase , atol=1e-3 ) )
# verify image_id
UpperCamelCase_ = torch.tensor([39769] )
self.assertTrue(torch.allclose(encoding['labels'][0]['image_id'] , _UpperCAmelCase ) )
# verify is_crowd
UpperCamelCase_ = torch.tensor([0, 0, 0, 0, 0, 0] )
self.assertTrue(torch.allclose(encoding['labels'][0]['iscrowd'] , _UpperCAmelCase ) )
# verify class_labels
UpperCamelCase_ = torch.tensor([75, 75, 63, 65, 17, 17] )
self.assertTrue(torch.allclose(encoding['labels'][0]['class_labels'] , _UpperCAmelCase ) )
# verify orig_size
UpperCamelCase_ = torch.tensor([480, 640] )
self.assertTrue(torch.allclose(encoding['labels'][0]['orig_size'] , _UpperCAmelCase ) )
# verify size
UpperCamelCase_ = torch.tensor([800, 1066] )
self.assertTrue(torch.allclose(encoding['labels'][0]['size'] , _UpperCAmelCase ) )
@slow
def _UpperCAmelCase ( self ) -> List[Any]:
# prepare image, target and masks_path
UpperCamelCase_ = Image.open('./tests/fixtures/tests_samples/COCO/000000039769.png' )
with open('./tests/fixtures/tests_samples/COCO/coco_panoptic_annotations.txt' , 'r' ) as f:
UpperCamelCase_ = json.loads(f.read() )
UpperCamelCase_ = {'file_name': '000000039769.png', 'image_id': 39769, 'segments_info': target}
UpperCamelCase_ = pathlib.Path('./tests/fixtures/tests_samples/COCO/coco_panoptic' )
# encode them
UpperCamelCase_ = ConditionalDetrImageProcessor(format='coco_panoptic' )
UpperCamelCase_ = image_processing(images=_UpperCAmelCase , annotations=_UpperCAmelCase , masks_path=_UpperCAmelCase , return_tensors='pt' )
# verify pixel values
UpperCamelCase_ = torch.Size([1, 3, 800, 1066] )
self.assertEqual(encoding['pixel_values'].shape , _UpperCAmelCase )
UpperCamelCase_ = torch.tensor([0.2_7_9_6, 0.3_1_3_8, 0.3_4_8_1] )
self.assertTrue(torch.allclose(encoding['pixel_values'][0, 0, 0, :3] , _UpperCAmelCase , atol=1e-4 ) )
# verify area
UpperCamelCase_ = torch.tensor([1_4_7_9_7_9.6_8_7_5, 1_6_5_5_2_7.0_4_6_9, 4_8_4_6_3_8.5_9_3_8, 1_1_2_9_2.9_3_7_5, 5_8_7_9.6_5_6_2, 7_6_3_4.1_1_4_7] )
self.assertTrue(torch.allclose(encoding['labels'][0]['area'] , _UpperCAmelCase ) )
# verify boxes
UpperCamelCase_ = torch.Size([6, 4] )
self.assertEqual(encoding['labels'][0]['boxes'].shape , _UpperCAmelCase )
UpperCamelCase_ = torch.tensor([0.2_6_2_5, 0.5_4_3_7, 0.4_6_8_8, 0.8_6_2_5] )
self.assertTrue(torch.allclose(encoding['labels'][0]['boxes'][0] , _UpperCAmelCase , atol=1e-3 ) )
# verify image_id
UpperCamelCase_ = torch.tensor([39769] )
self.assertTrue(torch.allclose(encoding['labels'][0]['image_id'] , _UpperCAmelCase ) )
# verify is_crowd
UpperCamelCase_ = torch.tensor([0, 0, 0, 0, 0, 0] )
self.assertTrue(torch.allclose(encoding['labels'][0]['iscrowd'] , _UpperCAmelCase ) )
# verify class_labels
UpperCamelCase_ = torch.tensor([17, 17, 63, 75, 75, 93] )
self.assertTrue(torch.allclose(encoding['labels'][0]['class_labels'] , _UpperCAmelCase ) )
# verify masks
UpperCamelCase_ = 822873
self.assertEqual(encoding['labels'][0]['masks'].sum().item() , _UpperCAmelCase )
# verify orig_size
UpperCamelCase_ = torch.tensor([480, 640] )
self.assertTrue(torch.allclose(encoding['labels'][0]['orig_size'] , _UpperCAmelCase ) )
# verify size
UpperCamelCase_ = torch.tensor([800, 1066] )
self.assertTrue(torch.allclose(encoding['labels'][0]['size'] , _UpperCAmelCase ) )
| 23 |
import argparse
import os
import evaluate
import torch
from datasets import load_dataset
from torch.optim import AdamW
from torch.utils.data import DataLoader
from transformers import AutoModelForSequenceClassification, AutoTokenizer, get_linear_schedule_with_warmup, set_seed
from accelerate import Accelerator, DistributedType
########################################################################
# This is a fully working simple example to use Accelerate
# and perform gradient accumulation
#
# This example trains a Bert base model on GLUE MRPC
# in any of the following settings (with the same script):
# - single CPU or single GPU
# - multi GPUS (using PyTorch distributed mode)
# - (multi) TPUs
# - fp16 (mixed-precision) or fp32 (normal precision)
#
# To run it in each of these various modes, follow the instructions
# in the readme for examples:
# https://github.com/huggingface/accelerate/tree/main/examples
#
########################################################################
snake_case__ : Dict = 1_6
snake_case__ : List[str] = 3_2
def _snake_case (__lowercase , __lowercase = 16):
UpperCamelCase_ = AutoTokenizer.from_pretrained('bert-base-cased')
UpperCamelCase_ = load_dataset('glue' , 'mrpc')
def tokenize_function(__lowercase):
# max_length=None => use the model max length (it's actually the default)
UpperCamelCase_ = tokenizer(examples['sentence1'] , examples['sentence2'] , truncation=__lowercase , max_length=__lowercase)
return outputs
# Apply the method we just defined to all the examples in all the splits of the dataset
# starting with the main process first:
with accelerator.main_process_first():
UpperCamelCase_ = datasets.map(
__lowercase , batched=__lowercase , remove_columns=['idx', 'sentence1', 'sentence2'] , )
# We also rename the 'label' column to 'labels' which is the expected name for labels by the models of the
# transformers library
UpperCamelCase_ = tokenized_datasets.rename_column('label' , 'labels')
def collate_fn(__lowercase):
# On TPU it's best to pad everything to the same length or training will be very slow.
UpperCamelCase_ = 128 if accelerator.distributed_type == DistributedType.TPU else None
# When using mixed precision we want round multiples of 8/16
if accelerator.mixed_precision == "fp8":
UpperCamelCase_ = 16
elif accelerator.mixed_precision != "no":
UpperCamelCase_ = 8
else:
UpperCamelCase_ = None
return tokenizer.pad(
__lowercase , padding='longest' , max_length=__lowercase , pad_to_multiple_of=__lowercase , return_tensors='pt' , )
# Instantiate dataloaders.
UpperCamelCase_ = DataLoader(
tokenized_datasets['train'] , shuffle=__lowercase , collate_fn=__lowercase , batch_size=__lowercase)
UpperCamelCase_ = DataLoader(
tokenized_datasets['validation'] , shuffle=__lowercase , collate_fn=__lowercase , batch_size=__lowercase)
return train_dataloader, eval_dataloader
# For testing only
if os.environ.get("""TESTING_MOCKED_DATALOADERS""", None) == "1":
from accelerate.test_utils.training import mocked_dataloaders
snake_case__ : List[str] = mocked_dataloaders # noqa: F811
def _snake_case (__lowercase , __lowercase):
# For testing only
if os.environ.get('TESTING_MOCKED_DATALOADERS' , __lowercase) == "1":
UpperCamelCase_ = 2
# New Code #
UpperCamelCase_ = int(args.gradient_accumulation_steps)
# Initialize accelerator
UpperCamelCase_ = Accelerator(
cpu=args.cpu , mixed_precision=args.mixed_precision , gradient_accumulation_steps=__lowercase)
if accelerator.distributed_type == DistributedType.TPU and gradient_accumulation_steps > 1:
raise NotImplementedError(
'Gradient accumulation on TPUs is currently not supported. Pass `gradient_accumulation_steps=1`')
# Sample hyper-parameters for learning rate, batch size, seed and a few other HPs
UpperCamelCase_ = config['lr']
UpperCamelCase_ = int(config['num_epochs'])
UpperCamelCase_ = int(config['seed'])
UpperCamelCase_ = int(config['batch_size'])
UpperCamelCase_ = evaluate.load('glue' , 'mrpc')
set_seed(__lowercase)
UpperCamelCase_ , UpperCamelCase_ = get_dataloaders(__lowercase , __lowercase)
# Instantiate the model (we build the model here so that the seed also control new weights initialization)
UpperCamelCase_ = AutoModelForSequenceClassification.from_pretrained('bert-base-cased' , return_dict=__lowercase)
# We could avoid this line since the accelerator is set with `device_placement=True` (default value).
# Note that if you are placing tensors on devices manually, this line absolutely needs to be before the optimizer
# creation otherwise training will not work on TPU (`accelerate` will kindly throw an error to make us aware of that).
UpperCamelCase_ = model.to(accelerator.device)
# Instantiate optimizer
UpperCamelCase_ = AdamW(params=model.parameters() , lr=__lowercase)
# Instantiate scheduler
UpperCamelCase_ = get_linear_schedule_with_warmup(
optimizer=__lowercase , num_warmup_steps=100 , num_training_steps=(len(__lowercase) * num_epochs) , )
# Prepare everything
# There is no specific order to remember, we just need to unpack the objects in the same order we gave them to the
# prepare method.
UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ = accelerator.prepare(
__lowercase , __lowercase , __lowercase , __lowercase , __lowercase)
# Now we train the model
for epoch in range(__lowercase):
model.train()
for step, batch in enumerate(__lowercase):
# We could avoid this line since we set the accelerator with `device_placement=True`.
batch.to(accelerator.device)
# New code #
# We use the new `accumulate` context manager to perform gradient accumulation
# We also currently do not support TPUs nor advise it as bugs were found on the XLA side when running our tests.
with accelerator.accumulate(__lowercase):
UpperCamelCase_ = model(**__lowercase)
UpperCamelCase_ = output.loss
accelerator.backward(__lowercase)
optimizer.step()
lr_scheduler.step()
optimizer.zero_grad()
model.eval()
for step, batch in enumerate(__lowercase):
# We could avoid this line since we set the accelerator with `device_placement=True`.
batch.to(accelerator.device)
with torch.no_grad():
UpperCamelCase_ = model(**__lowercase)
UpperCamelCase_ = outputs.logits.argmax(dim=-1)
UpperCamelCase_ , UpperCamelCase_ = accelerator.gather_for_metrics((predictions, batch['labels']))
metric.add_batch(
predictions=__lowercase , references=__lowercase , )
UpperCamelCase_ = metric.compute()
# Use accelerator.print to print only on the main process.
accelerator.print(f"""epoch {epoch}:""" , __lowercase)
def _snake_case ():
UpperCamelCase_ = argparse.ArgumentParser(description='Simple example of training script.')
parser.add_argument(
'--mixed_precision' , type=__lowercase , default=__lowercase , choices=['no', 'fp16', 'bf16', 'fp8'] , help='Whether to use mixed precision. Choose'
'between fp16 and bf16 (bfloat16). Bf16 requires PyTorch >= 1.10.'
'and an Nvidia Ampere GPU.' , )
# New Code #
parser.add_argument(
'--gradient_accumulation_steps' , type=__lowercase , default=1 , help='The number of minibatches to be ran before gradients are accumulated.' , )
parser.add_argument('--cpu' , action='store_true' , help='If passed, will train on the CPU.')
UpperCamelCase_ = parser.parse_args()
UpperCamelCase_ = {'lr': 2e-5, 'num_epochs': 3, 'seed': 42, 'batch_size': 16}
training_function(__lowercase , __lowercase)
if __name__ == "__main__":
main()
| 23 | 1 |
import unittest
import numpy as np
import torch
from transformers import CLIPTextConfig, CLIPTextModel, CLIPTokenizer
from diffusers import (
AutoencoderKL,
DDIMScheduler,
DPMSolverMultistepScheduler,
TextToVideoSDPipeline,
UNetaDConditionModel,
)
from diffusers.utils import is_xformers_available, load_numpy, skip_mps, slow, torch_device
from diffusers.utils.testing_utils import enable_full_determinism
from ..pipeline_params import TEXT_TO_IMAGE_BATCH_PARAMS, TEXT_TO_IMAGE_PARAMS
from ..test_pipelines_common import PipelineTesterMixin
enable_full_determinism()
@skip_mps
class _a ( UpperCAmelCase__ , unittest.TestCase ):
"""simple docstring"""
A_ = TextToVideoSDPipeline
A_ = TEXT_TO_IMAGE_PARAMS
A_ = TEXT_TO_IMAGE_BATCH_PARAMS
# No `output_type`.
A_ = frozenset(
[
"""num_inference_steps""",
"""generator""",
"""latents""",
"""return_dict""",
"""callback""",
"""callback_steps""",
] )
def _UpperCAmelCase ( self ) -> List[str]:
torch.manual_seed(0 )
UpperCamelCase_ = UNetaDConditionModel(
block_out_channels=(32, 64, 64, 64) , layers_per_block=2 , sample_size=32 , in_channels=4 , out_channels=4 , down_block_types=('CrossAttnDownBlock3D', 'CrossAttnDownBlock3D', 'CrossAttnDownBlock3D', 'DownBlock3D') , up_block_types=('UpBlock3D', 'CrossAttnUpBlock3D', 'CrossAttnUpBlock3D', 'CrossAttnUpBlock3D') , cross_attention_dim=32 , attention_head_dim=4 , )
UpperCamelCase_ = DDIMScheduler(
beta_start=0.0_0_0_8_5 , beta_end=0.0_1_2 , beta_schedule='scaled_linear' , clip_sample=_UpperCAmelCase , set_alpha_to_one=_UpperCAmelCase , )
torch.manual_seed(0 )
UpperCamelCase_ = AutoencoderKL(
block_out_channels=[32, 64] , in_channels=3 , out_channels=3 , down_block_types=['DownEncoderBlock2D', 'DownEncoderBlock2D'] , up_block_types=['UpDecoderBlock2D', 'UpDecoderBlock2D'] , latent_channels=4 , sample_size=128 , )
torch.manual_seed(0 )
UpperCamelCase_ = CLIPTextConfig(
bos_token_id=0 , eos_token_id=2 , hidden_size=32 , intermediate_size=37 , layer_norm_eps=1e-05 , num_attention_heads=4 , num_hidden_layers=5 , pad_token_id=1 , vocab_size=1000 , hidden_act='gelu' , projection_dim=512 , )
UpperCamelCase_ = CLIPTextModel(_UpperCAmelCase )
UpperCamelCase_ = CLIPTokenizer.from_pretrained('hf-internal-testing/tiny-random-clip' )
UpperCamelCase_ = {
'unet': unet,
'scheduler': scheduler,
'vae': vae,
'text_encoder': text_encoder,
'tokenizer': tokenizer,
}
return components
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase=0 ) -> Optional[Any]:
if str(_UpperCAmelCase ).startswith('mps' ):
UpperCamelCase_ = torch.manual_seed(_UpperCAmelCase )
else:
UpperCamelCase_ = torch.Generator(device=_UpperCAmelCase ).manual_seed(_UpperCAmelCase )
UpperCamelCase_ = {
'prompt': 'A painting of a squirrel eating a burger',
'generator': generator,
'num_inference_steps': 2,
'guidance_scale': 6.0,
'output_type': 'pt',
}
return inputs
def _UpperCAmelCase ( self ) -> List[Any]:
UpperCamelCase_ = 'cpu' # ensure determinism for the device-dependent torch.Generator
UpperCamelCase_ = self.get_dummy_components()
UpperCamelCase_ = TextToVideoSDPipeline(**_UpperCAmelCase )
UpperCamelCase_ = sd_pipe.to(_UpperCAmelCase )
sd_pipe.set_progress_bar_config(disable=_UpperCAmelCase )
UpperCamelCase_ = self.get_dummy_inputs(_UpperCAmelCase )
UpperCamelCase_ = 'np'
UpperCamelCase_ = sd_pipe(**_UpperCAmelCase ).frames
UpperCamelCase_ = frames[0][-3:, -3:, -1]
assert frames[0].shape == (64, 64, 3)
UpperCamelCase_ = np.array([1_5_8.0, 1_6_0.0, 1_5_3.0, 1_2_5.0, 1_0_0.0, 1_2_1.0, 1_1_1.0, 9_3.0, 1_1_3.0] )
assert np.abs(image_slice.flatten() - expected_slice ).max() < 1e-2
def _UpperCAmelCase ( self ) -> Any:
self._test_attention_slicing_forward_pass(test_mean_pixel_difference=_UpperCAmelCase , expected_max_diff=3e-3 )
@unittest.skipIf(
torch_device != 'cuda' or not is_xformers_available() , reason='XFormers attention is only available with CUDA and `xformers` installed' , )
def _UpperCAmelCase ( self ) -> List[Any]:
self._test_xformers_attention_forwardGenerator_pass(test_mean_pixel_difference=_UpperCAmelCase , expected_max_diff=1e-2 )
@unittest.skip(reason='Batching needs to be properly figured out first for this pipeline.' )
def _UpperCAmelCase ( self ) -> Optional[int]:
pass
@unittest.skip(reason='Batching needs to be properly figured out first for this pipeline.' )
def _UpperCAmelCase ( self ) -> List[Any]:
pass
@unittest.skip(reason='`num_images_per_prompt` argument is not supported for this pipeline.' )
def _UpperCAmelCase ( self ) -> Dict:
pass
def _UpperCAmelCase ( self ) -> Tuple:
return super().test_progress_bar()
@slow
@skip_mps
class _a ( unittest.TestCase ):
"""simple docstring"""
def _UpperCAmelCase ( self ) -> Union[str, Any]:
UpperCamelCase_ = load_numpy(
'https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/text_to_video/video.npy' )
UpperCamelCase_ = TextToVideoSDPipeline.from_pretrained('damo-vilab/text-to-video-ms-1.7b' )
UpperCamelCase_ = DPMSolverMultistepScheduler.from_config(pipe.scheduler.config )
UpperCamelCase_ = pipe.to('cuda' )
UpperCamelCase_ = 'Spiderman is surfing'
UpperCamelCase_ = torch.Generator(device='cpu' ).manual_seed(0 )
UpperCamelCase_ = pipe(_UpperCAmelCase , generator=_UpperCAmelCase , num_inference_steps=25 , output_type='pt' ).frames
UpperCamelCase_ = video_frames.cpu().numpy()
assert np.abs(expected_video - video ).mean() < 5e-2
def _UpperCAmelCase ( self ) -> List[Any]:
UpperCamelCase_ = load_numpy(
'https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/text_to_video/video_2step.npy' )
UpperCamelCase_ = TextToVideoSDPipeline.from_pretrained('damo-vilab/text-to-video-ms-1.7b' )
UpperCamelCase_ = pipe.to('cuda' )
UpperCamelCase_ = 'Spiderman is surfing'
UpperCamelCase_ = torch.Generator(device='cpu' ).manual_seed(0 )
UpperCamelCase_ = pipe(_UpperCAmelCase , generator=_UpperCAmelCase , num_inference_steps=2 , output_type='pt' ).frames
UpperCamelCase_ = video_frames.cpu().numpy()
assert np.abs(expected_video - video ).mean() < 5e-2
| 23 |
import inspect
import unittest
import numpy as np
from tests.test_modeling_common import floats_tensor
from transformers import MaskaFormerConfig, is_torch_available, is_vision_available
from transformers.testing_utils import require_torch, require_torch_multi_gpu, require_vision, slow, torch_device
from transformers.utils import cached_property
from ...test_configuration_common import ConfigTester
from ...test_modeling_common import ModelTesterMixin
from ...test_pipeline_mixin import PipelineTesterMixin
if is_torch_available():
import torch
from transformers import MaskaFormerForUniversalSegmentation, MaskaFormerModel
if is_vision_available():
from transformers import MaskaFormerImageProcessor
if is_vision_available():
from PIL import Image
class _a :
"""simple docstring"""
def __init__( self , _UpperCAmelCase , _UpperCAmelCase=2 , _UpperCAmelCase=True , _UpperCAmelCase=False , _UpperCAmelCase=10 , _UpperCAmelCase=3 , _UpperCAmelCase=32 * 8 , _UpperCAmelCase=32 * 8 , _UpperCAmelCase=4 , _UpperCAmelCase=64 , ) -> List[Any]:
UpperCamelCase_ = parent
UpperCamelCase_ = batch_size
UpperCamelCase_ = is_training
UpperCamelCase_ = use_auxiliary_loss
UpperCamelCase_ = num_queries
UpperCamelCase_ = num_channels
UpperCamelCase_ = min_size
UpperCamelCase_ = max_size
UpperCamelCase_ = num_labels
UpperCamelCase_ = hidden_dim
UpperCamelCase_ = hidden_dim
def _UpperCAmelCase ( self ) -> List[str]:
UpperCamelCase_ = floats_tensor([self.batch_size, self.num_channels, self.min_size, self.max_size] ).to(
_UpperCAmelCase )
UpperCamelCase_ = torch.ones([self.batch_size, self.min_size, self.max_size] , device=_UpperCAmelCase )
UpperCamelCase_ = (
torch.rand([self.batch_size, self.num_labels, self.min_size, self.max_size] , device=_UpperCAmelCase ) > 0.5
).float()
UpperCamelCase_ = (torch.rand((self.batch_size, self.num_labels) , device=_UpperCAmelCase ) > 0.5).long()
UpperCamelCase_ = self.get_config()
return config, pixel_values, pixel_mask, mask_labels, class_labels
def _UpperCAmelCase ( self ) -> Optional[int]:
UpperCamelCase_ = MaskaFormerConfig(
hidden_size=self.hidden_dim , )
UpperCamelCase_ = self.num_queries
UpperCamelCase_ = self.num_labels
UpperCamelCase_ = [1, 1, 1, 1]
UpperCamelCase_ = self.num_channels
UpperCamelCase_ = 64
UpperCamelCase_ = 128
UpperCamelCase_ = self.hidden_dim
UpperCamelCase_ = self.hidden_dim
UpperCamelCase_ = self.hidden_dim
return config
def _UpperCAmelCase ( self ) -> Dict:
UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ = self.prepare_config_and_inputs()
UpperCamelCase_ = {'pixel_values': pixel_values, 'pixel_mask': pixel_mask}
return config, inputs_dict
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> List[Any]:
UpperCamelCase_ = output.encoder_hidden_states
UpperCamelCase_ = output.pixel_decoder_hidden_states
UpperCamelCase_ = output.transformer_decoder_hidden_states
self.parent.assertTrue(len(_UpperCAmelCase ) , len(config.backbone_config.depths ) )
self.parent.assertTrue(len(_UpperCAmelCase ) , len(config.backbone_config.depths ) )
self.parent.assertTrue(len(_UpperCAmelCase ) , config.decoder_layers )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase=False ) -> Any:
with torch.no_grad():
UpperCamelCase_ = MaskaFormerModel(config=_UpperCAmelCase )
model.to(_UpperCAmelCase )
model.eval()
UpperCamelCase_ = model(pixel_values=_UpperCAmelCase , pixel_mask=_UpperCAmelCase )
UpperCamelCase_ = model(_UpperCAmelCase , output_hidden_states=_UpperCAmelCase )
self.parent.assertEqual(
output.transformer_decoder_last_hidden_state.shape , (self.batch_size, self.num_queries, self.hidden_dim) , )
# let's ensure the other two hidden state exists
self.parent.assertTrue(output.pixel_decoder_last_hidden_state is not None )
self.parent.assertTrue(output.encoder_last_hidden_state is not None )
if output_hidden_states:
self.check_output_hidden_state(_UpperCAmelCase , _UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase ) -> Union[str, Any]:
UpperCamelCase_ = MaskaFormerForUniversalSegmentation(config=_UpperCAmelCase )
model.to(_UpperCAmelCase )
model.eval()
def comm_check_on_output(_UpperCAmelCase ):
# let's still check that all the required stuff is there
self.parent.assertTrue(result.transformer_decoder_last_hidden_state is not None )
self.parent.assertTrue(result.pixel_decoder_last_hidden_state is not None )
self.parent.assertTrue(result.encoder_last_hidden_state is not None )
# okay, now we need to check the logits shape
# due to the encoder compression, masks have a //4 spatial size
self.parent.assertEqual(
result.masks_queries_logits.shape , (self.batch_size, self.num_queries, self.min_size // 4, self.max_size // 4) , )
# + 1 for null class
self.parent.assertEqual(
result.class_queries_logits.shape , (self.batch_size, self.num_queries, self.num_labels + 1) )
with torch.no_grad():
UpperCamelCase_ = model(pixel_values=_UpperCAmelCase , pixel_mask=_UpperCAmelCase )
UpperCamelCase_ = model(_UpperCAmelCase )
comm_check_on_output(_UpperCAmelCase )
UpperCamelCase_ = model(
pixel_values=_UpperCAmelCase , pixel_mask=_UpperCAmelCase , mask_labels=_UpperCAmelCase , class_labels=_UpperCAmelCase )
comm_check_on_output(_UpperCAmelCase )
self.parent.assertTrue(result.loss is not None )
self.parent.assertEqual(result.loss.shape , torch.Size([1] ) )
@require_torch
class _a ( UpperCAmelCase__ , UpperCAmelCase__ , unittest.TestCase ):
"""simple docstring"""
A_ = (MaskaFormerModel, MaskaFormerForUniversalSegmentation) if is_torch_available() else ()
A_ = {"""feature-extraction""": MaskaFormerModel} if is_torch_available() else {}
A_ = False
A_ = False
A_ = False
A_ = False
def _UpperCAmelCase ( self ) -> Optional[Any]:
UpperCamelCase_ = MaskaFormerModelTester(self )
UpperCamelCase_ = ConfigTester(self , config_class=_UpperCAmelCase , has_text_modality=_UpperCAmelCase )
def _UpperCAmelCase ( self ) -> Union[str, Any]:
self.config_tester.run_common_tests()
def _UpperCAmelCase ( self ) -> Union[str, Any]:
UpperCamelCase_ , UpperCamelCase_ = self.model_tester.prepare_config_and_inputs_for_common()
self.model_tester.create_and_check_maskaformer_model(_UpperCAmelCase , **_UpperCAmelCase , output_hidden_states=_UpperCAmelCase )
def _UpperCAmelCase ( self ) -> List[Any]:
UpperCamelCase_ = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_maskaformer_instance_segmentation_head_model(*_UpperCAmelCase )
@unittest.skip(reason='Mask2Former does not use inputs_embeds' )
def _UpperCAmelCase ( self ) -> Any:
pass
@unittest.skip(reason='Mask2Former does not have a get_input_embeddings method' )
def _UpperCAmelCase ( self ) -> Optional[int]:
pass
@unittest.skip(reason='Mask2Former is not a generative model' )
def _UpperCAmelCase ( self ) -> Any:
pass
@unittest.skip(reason='Mask2Former does not use token embeddings' )
def _UpperCAmelCase ( self ) -> Optional[Any]:
pass
@require_torch_multi_gpu
@unittest.skip(
reason='Mask2Former has some layers using `add_module` which doesn\'t work well with `nn.DataParallel`' )
def _UpperCAmelCase ( self ) -> int:
pass
@unittest.skip('Will be fixed soon by reducing the size of the model used for common tests.' )
def _UpperCAmelCase ( self ) -> str:
pass
def _UpperCAmelCase ( self ) -> Optional[int]:
UpperCamelCase_ , UpperCamelCase_ = self.model_tester.prepare_config_and_inputs_for_common()
for model_class in self.all_model_classes:
UpperCamelCase_ = model_class(_UpperCAmelCase )
UpperCamelCase_ = inspect.signature(model.forward )
# signature.parameters is an OrderedDict => so arg_names order is deterministic
UpperCamelCase_ = [*signature.parameters.keys()]
UpperCamelCase_ = ['pixel_values']
self.assertListEqual(arg_names[:1] , _UpperCAmelCase )
@slow
def _UpperCAmelCase ( self ) -> Tuple:
for model_name in ["facebook/mask2former-swin-small-coco-instance"]:
UpperCamelCase_ = MaskaFormerModel.from_pretrained(_UpperCAmelCase )
self.assertIsNotNone(_UpperCAmelCase )
def _UpperCAmelCase ( self ) -> Dict:
UpperCamelCase_ = (self.model_tester.min_size,) * 2
UpperCamelCase_ = {
'pixel_values': torch.randn((2, 3, *size) , device=_UpperCAmelCase ),
'mask_labels': torch.randn((2, 10, *size) , device=_UpperCAmelCase ),
'class_labels': torch.zeros(2 , 10 , device=_UpperCAmelCase ).long(),
}
UpperCamelCase_ = self.model_tester.get_config()
UpperCamelCase_ = MaskaFormerForUniversalSegmentation(_UpperCAmelCase ).to(_UpperCAmelCase )
UpperCamelCase_ = model(**_UpperCAmelCase )
self.assertTrue(outputs.loss is not None )
def _UpperCAmelCase ( self ) -> str:
UpperCamelCase_ , UpperCamelCase_ = self.model_tester.prepare_config_and_inputs_for_common()
self.model_tester.create_and_check_maskaformer_model(_UpperCAmelCase , **_UpperCAmelCase , output_hidden_states=_UpperCAmelCase )
def _UpperCAmelCase ( self ) -> Optional[int]:
UpperCamelCase_ , UpperCamelCase_ = self.model_tester.prepare_config_and_inputs_for_common()
for model_class in self.all_model_classes:
UpperCamelCase_ = model_class(_UpperCAmelCase ).to(_UpperCAmelCase )
UpperCamelCase_ = model(**_UpperCAmelCase , output_attentions=_UpperCAmelCase )
self.assertTrue(outputs.attentions is not None )
def _UpperCAmelCase ( self ) -> List[Any]:
if not self.model_tester.is_training:
return
UpperCamelCase_ = self.all_model_classes[1]
UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ = self.model_tester.prepare_config_and_inputs()
UpperCamelCase_ = model_class(_UpperCAmelCase )
model.to(_UpperCAmelCase )
model.train()
UpperCamelCase_ = model(_UpperCAmelCase , mask_labels=_UpperCAmelCase , class_labels=_UpperCAmelCase ).loss
loss.backward()
def _UpperCAmelCase ( self ) -> int:
UpperCamelCase_ = self.all_model_classes[1]
UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ = self.model_tester.prepare_config_and_inputs()
UpperCamelCase_ = True
UpperCamelCase_ = True
UpperCamelCase_ = model_class(_UpperCAmelCase ).to(_UpperCAmelCase )
model.train()
UpperCamelCase_ = model(_UpperCAmelCase , mask_labels=_UpperCAmelCase , class_labels=_UpperCAmelCase )
UpperCamelCase_ = outputs.encoder_hidden_states[0]
encoder_hidden_states.retain_grad()
UpperCamelCase_ = outputs.pixel_decoder_hidden_states[0]
pixel_decoder_hidden_states.retain_grad()
UpperCamelCase_ = outputs.transformer_decoder_hidden_states[0]
transformer_decoder_hidden_states.retain_grad()
UpperCamelCase_ = outputs.attentions[0]
attentions.retain_grad()
outputs.loss.backward(retain_graph=_UpperCAmelCase )
self.assertIsNotNone(encoder_hidden_states.grad )
self.assertIsNotNone(pixel_decoder_hidden_states.grad )
self.assertIsNotNone(transformer_decoder_hidden_states.grad )
self.assertIsNotNone(attentions.grad )
snake_case__ : List[Any] = 1E-4
def _snake_case ():
UpperCamelCase_ = Image.open('./tests/fixtures/tests_samples/COCO/000000039769.png')
return image
@require_vision
@slow
class _a ( unittest.TestCase ):
"""simple docstring"""
@cached_property
def _UpperCAmelCase ( self ) -> Optional[int]:
return "facebook/mask2former-swin-small-coco-instance"
@cached_property
def _UpperCAmelCase ( self ) -> List[str]:
return MaskaFormerImageProcessor.from_pretrained(self.model_checkpoints ) if is_vision_available() else None
def _UpperCAmelCase ( self ) -> str:
UpperCamelCase_ = MaskaFormerModel.from_pretrained(self.model_checkpoints ).to(_UpperCAmelCase )
UpperCamelCase_ = self.default_image_processor
UpperCamelCase_ = prepare_img()
UpperCamelCase_ = image_processor(_UpperCAmelCase , return_tensors='pt' ).to(_UpperCAmelCase )
UpperCamelCase_ = inputs['pixel_values'].shape
# check size is divisible by 32
self.assertTrue((inputs_shape[-1] % 32) == 0 and (inputs_shape[-2] % 32) == 0 )
# check size
self.assertEqual(_UpperCAmelCase , (1, 3, 384, 384) )
with torch.no_grad():
UpperCamelCase_ = model(**_UpperCAmelCase )
UpperCamelCase_ = torch.tensor(
[[-0.2_7_9_0, -1.0_7_1_7, -1.1_6_6_8], [-0.5_1_2_8, -0.3_1_2_8, -0.4_9_8_7], [-0.5_8_3_2, 0.1_9_7_1, -0.0_1_9_7]] ).to(_UpperCAmelCase )
self.assertTrue(
torch.allclose(
outputs.encoder_last_hidden_state[0, 0, :3, :3] , _UpperCAmelCase , atol=_UpperCAmelCase ) )
UpperCamelCase_ = torch.tensor(
[[0.8_9_7_3, 1.1_8_4_7, 1.1_7_7_6], [1.1_9_3_4, 1.5_0_4_0, 1.5_1_2_8], [1.1_1_5_3, 1.4_4_8_6, 1.4_9_5_1]] ).to(_UpperCAmelCase )
self.assertTrue(
torch.allclose(
outputs.pixel_decoder_last_hidden_state[0, 0, :3, :3] , _UpperCAmelCase , atol=_UpperCAmelCase ) )
UpperCamelCase_ = torch.tensor(
[[2.1_1_5_2, 1.7_0_0_0, -0.8_6_0_3], [1.5_8_0_8, 1.8_0_0_4, -0.9_3_5_3], [1.6_0_4_3, 1.7_4_9_5, -0.5_9_9_9]] ).to(_UpperCAmelCase )
self.assertTrue(
torch.allclose(
outputs.transformer_decoder_last_hidden_state[0, :3, :3] , _UpperCAmelCase , atol=_UpperCAmelCase ) )
def _UpperCAmelCase ( self ) -> Optional[Any]:
UpperCamelCase_ = MaskaFormerForUniversalSegmentation.from_pretrained(self.model_checkpoints ).to(_UpperCAmelCase ).eval()
UpperCamelCase_ = self.default_image_processor
UpperCamelCase_ = prepare_img()
UpperCamelCase_ = image_processor(_UpperCAmelCase , return_tensors='pt' ).to(_UpperCAmelCase )
UpperCamelCase_ = inputs['pixel_values'].shape
# check size is divisible by 32
self.assertTrue((inputs_shape[-1] % 32) == 0 and (inputs_shape[-2] % 32) == 0 )
# check size
self.assertEqual(_UpperCAmelCase , (1, 3, 384, 384) )
with torch.no_grad():
UpperCamelCase_ = model(**_UpperCAmelCase )
# masks_queries_logits
UpperCamelCase_ = outputs.masks_queries_logits
self.assertEqual(
masks_queries_logits.shape , (1, model.config.num_queries, inputs_shape[-2] // 4, inputs_shape[-1] // 4) )
UpperCamelCase_ = [
[-8.7_8_3_9, -9.0_0_5_6, -8.8_1_2_1],
[-7.4_1_0_4, -7.0_3_1_3, -6.5_4_0_1],
[-6.6_1_0_5, -6.3_4_2_7, -6.4_6_7_5],
]
UpperCamelCase_ = torch.tensor(_UpperCAmelCase ).to(_UpperCAmelCase )
self.assertTrue(torch.allclose(masks_queries_logits[0, 0, :3, :3] , _UpperCAmelCase , atol=_UpperCAmelCase ) )
# class_queries_logits
UpperCamelCase_ = outputs.class_queries_logits
self.assertEqual(class_queries_logits.shape , (1, model.config.num_queries, model.config.num_labels + 1) )
UpperCamelCase_ = torch.tensor(
[
[1.8_3_2_4, -8.0_8_3_5, -4.1_9_2_2],
[0.8_4_5_0, -9.0_0_5_0, -3.6_0_5_3],
[0.3_0_4_5, -7.7_2_9_3, -3.0_2_7_5],
] ).to(_UpperCAmelCase )
self.assertTrue(torch.allclose(outputs.class_queries_logits[0, :3, :3] , _UpperCAmelCase , atol=_UpperCAmelCase ) )
def _UpperCAmelCase ( self ) -> Dict:
UpperCamelCase_ = MaskaFormerForUniversalSegmentation.from_pretrained(self.model_checkpoints ).to(_UpperCAmelCase ).eval()
UpperCamelCase_ = self.default_image_processor
UpperCamelCase_ = image_processor(
[np.zeros((3, 800, 1333) ), np.zeros((3, 800, 1333) )] , segmentation_maps=[np.zeros((384, 384) ).astype(np.floataa ), np.zeros((384, 384) ).astype(np.floataa )] , return_tensors='pt' , )
UpperCamelCase_ = inputs['pixel_values'].to(_UpperCAmelCase )
UpperCamelCase_ = [el.to(_UpperCAmelCase ) for el in inputs['mask_labels']]
UpperCamelCase_ = [el.to(_UpperCAmelCase ) for el in inputs['class_labels']]
with torch.no_grad():
UpperCamelCase_ = model(**_UpperCAmelCase )
self.assertTrue(outputs.loss is not None )
| 23 | 1 |
import math
def _snake_case (__lowercase):
UpperCamelCase_ = []
UpperCamelCase_ = 2
UpperCamelCase_ = int(math.sqrt(__lowercase)) # Size of every segment
UpperCamelCase_ = [True] * (end + 1)
UpperCamelCase_ = []
while start <= end:
if temp[start] is True:
in_prime.append(__lowercase)
for i in range(start * start , end + 1 , __lowercase):
UpperCamelCase_ = False
start += 1
prime += in_prime
UpperCamelCase_ = end + 1
UpperCamelCase_ = min(2 * end , __lowercase)
while low <= n:
UpperCamelCase_ = [True] * (high - low + 1)
for each in in_prime:
UpperCamelCase_ = math.floor(low / each) * each
if t < low:
t += each
for j in range(__lowercase , high + 1 , __lowercase):
UpperCamelCase_ = False
for j in range(len(__lowercase)):
if temp[j] is True:
prime.append(j + low)
UpperCamelCase_ = high + 1
UpperCamelCase_ = min(high + end , __lowercase)
return prime
print(sieve(1_0**6))
| 23 |
import copy
from typing import TYPE_CHECKING, Any, Mapping, Optional, OrderedDict
from packaging import version
from ...configuration_utils import PretrainedConfig
from ...onnx import OnnxConfig
from ...utils import logging
from ..auto.configuration_auto import AutoConfig
if TYPE_CHECKING:
from ... import PreTrainedTokenizerBase, TensorType
snake_case__ : List[str] = logging.get_logger(__name__)
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
A_ = """vision-encoder-decoder"""
A_ = True
def __init__( self , **_UpperCAmelCase ) -> Dict:
super().__init__(**_UpperCAmelCase )
if "encoder" not in kwargs or "decoder" not in kwargs:
raise ValueError(
f"""A configuraton of type {self.model_type} cannot be instantiated because """
f"""not both `encoder` and `decoder` sub-configurations are passed, but only {kwargs}""" )
UpperCamelCase_ = kwargs.pop('encoder' )
UpperCamelCase_ = encoder_config.pop('model_type' )
UpperCamelCase_ = kwargs.pop('decoder' )
UpperCamelCase_ = decoder_config.pop('model_type' )
UpperCamelCase_ = AutoConfig.for_model(_UpperCAmelCase , **_UpperCAmelCase )
UpperCamelCase_ = AutoConfig.for_model(_UpperCAmelCase , **_UpperCAmelCase )
UpperCamelCase_ = True
@classmethod
def _UpperCAmelCase ( cls , _UpperCAmelCase , _UpperCAmelCase , **_UpperCAmelCase ) -> PretrainedConfig:
logger.info('Setting `config.is_decoder=True` and `config.add_cross_attention=True` for decoder_config' )
UpperCamelCase_ = True
UpperCamelCase_ = True
return cls(encoder=encoder_config.to_dict() , decoder=decoder_config.to_dict() , **_UpperCAmelCase )
def _UpperCAmelCase ( self ) -> int:
UpperCamelCase_ = copy.deepcopy(self.__dict__ )
UpperCamelCase_ = self.encoder.to_dict()
UpperCamelCase_ = self.decoder.to_dict()
UpperCamelCase_ = self.__class__.model_type
return output
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
A_ = version.parse("""1.11""" )
@property
def _UpperCAmelCase ( self ) -> Mapping[str, Mapping[int, str]]:
return OrderedDict(
[
('pixel_values', {0: 'batch', 1: 'num_channels', 2: 'height', 3: 'width'}),
] )
@property
def _UpperCAmelCase ( self ) -> float:
return 1e-4
@property
def _UpperCAmelCase ( self ) -> Mapping[str, Mapping[int, str]]:
return OrderedDict({'last_hidden_state': {0: 'batch', 1: 'encoder_sequence'}} )
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
@property
def _UpperCAmelCase ( self ) -> Mapping[str, Mapping[int, str]]:
UpperCamelCase_ = OrderedDict()
UpperCamelCase_ = {0: 'batch', 1: 'past_decoder_sequence + sequence'}
UpperCamelCase_ = {0: 'batch', 1: 'past_decoder_sequence + sequence'}
UpperCamelCase_ = {0: 'batch', 1: 'encoder_sequence'}
return common_inputs
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase = -1 , _UpperCAmelCase = -1 , _UpperCAmelCase = False , _UpperCAmelCase = None , ) -> Mapping[str, Any]:
import torch
UpperCamelCase_ = OrderedDict()
UpperCamelCase_ = super().generate_dummy_inputs(
_UpperCAmelCase , batch_size=_UpperCAmelCase , seq_length=_UpperCAmelCase , is_pair=_UpperCAmelCase , framework=_UpperCAmelCase )
UpperCamelCase_ , UpperCamelCase_ = dummy_input['input_ids'].shape
UpperCamelCase_ = (batch, encoder_sequence, self._config.encoder_hidden_size)
UpperCamelCase_ = dummy_input.pop('input_ids' )
UpperCamelCase_ = dummy_input.pop('attention_mask' )
UpperCamelCase_ = torch.zeros(_UpperCAmelCase )
return common_inputs
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
@property
def _UpperCAmelCase ( self ) -> None:
pass
def _UpperCAmelCase ( self , _UpperCAmelCase ) -> OnnxConfig:
return VisionEncoderDecoderEncoderOnnxConfig(_UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase = "default" ) -> OnnxConfig:
UpperCamelCase_ = encoder_config.hidden_size
return VisionEncoderDecoderDecoderOnnxConfig(_UpperCAmelCase , _UpperCAmelCase )
| 23 | 1 |
import torch
from torch import nn
from ...configuration_utils import ConfigMixin, register_to_config
from ...models import ModelMixin
class _a ( UpperCAmelCase__ , UpperCAmelCase__ ):
"""simple docstring"""
@register_to_config
def __init__( self , *,
_UpperCAmelCase = 4 , _UpperCAmelCase = 768 , _UpperCAmelCase , _UpperCAmelCase , ) -> Any:
super().__init__()
UpperCamelCase_ = nn.Parameter(torch.zeros(_UpperCAmelCase ) )
# parameters for additional clip time embeddings
UpperCamelCase_ = nn.Linear(_UpperCAmelCase , _UpperCAmelCase )
UpperCamelCase_ = nn.Linear(_UpperCAmelCase , _UpperCAmelCase )
# parameters for encoder hidden states
UpperCamelCase_ = clip_extra_context_tokens
UpperCamelCase_ = nn.Linear(
_UpperCAmelCase , self.clip_extra_context_tokens * cross_attention_dim )
UpperCamelCase_ = nn.Linear(_UpperCAmelCase , _UpperCAmelCase )
UpperCamelCase_ = nn.LayerNorm(_UpperCAmelCase )
def _UpperCAmelCase ( self , *, _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase ) -> Dict:
if do_classifier_free_guidance:
# Add the classifier free guidance embeddings to the image embeddings
UpperCamelCase_ = image_embeddings.shape[0]
UpperCamelCase_ = self.learned_classifier_free_guidance_embeddings.unsqueeze(0 )
UpperCamelCase_ = classifier_free_guidance_embeddings.expand(
_UpperCAmelCase , -1 )
UpperCamelCase_ = torch.cat([classifier_free_guidance_embeddings, image_embeddings] , dim=0 )
# The image embeddings batch size and the text embeddings batch size are equal
assert image_embeddings.shape[0] == prompt_embeds.shape[0]
UpperCamelCase_ = prompt_embeds.shape[0]
# "Specifically, we modify the architecture described in Nichol et al. (2021) by projecting and
# adding CLIP embeddings to the existing timestep embedding, ...
UpperCamelCase_ = self.embedding_proj(_UpperCAmelCase )
UpperCamelCase_ = self.clip_image_embeddings_project_to_time_embeddings(_UpperCAmelCase )
UpperCamelCase_ = time_projected_image_embeddings + time_projected_prompt_embeds
# ... and by projecting CLIP embeddings into four
# extra tokens of context that are concatenated to the sequence of outputs from the GLIDE text encoder"
UpperCamelCase_ = self.clip_extra_context_tokens_proj(_UpperCAmelCase )
UpperCamelCase_ = clip_extra_context_tokens.reshape(_UpperCAmelCase , -1 , self.clip_extra_context_tokens )
UpperCamelCase_ = clip_extra_context_tokens.permute(0 , 2 , 1 )
UpperCamelCase_ = self.encoder_hidden_states_proj(_UpperCAmelCase )
UpperCamelCase_ = self.text_encoder_hidden_states_norm(_UpperCAmelCase )
UpperCamelCase_ = torch.cat([clip_extra_context_tokens, text_encoder_hidden_states] , dim=1 )
return text_encoder_hidden_states, additive_clip_time_embeddings
| 23 |
import argparse
import torch
from transformers import MobileBertConfig, MobileBertForPreTraining, load_tf_weights_in_mobilebert
from transformers.utils import logging
logging.set_verbosity_info()
def _snake_case (__lowercase , __lowercase , __lowercase):
# Initialise PyTorch model
UpperCamelCase_ = MobileBertConfig.from_json_file(__lowercase)
print(f"""Building PyTorch model from configuration: {config}""")
UpperCamelCase_ = MobileBertForPreTraining(__lowercase)
# Load weights from tf checkpoint
UpperCamelCase_ = load_tf_weights_in_mobilebert(__lowercase , __lowercase , __lowercase)
# Save pytorch-model
print(f"""Save PyTorch model to {pytorch_dump_path}""")
torch.save(model.state_dict() , __lowercase)
if __name__ == "__main__":
snake_case__ : int = argparse.ArgumentParser()
# Required parameters
parser.add_argument(
"""--tf_checkpoint_path""", default=None, type=str, required=True, help="""Path to the TensorFlow checkpoint path."""
)
parser.add_argument(
"""--mobilebert_config_file""",
default=None,
type=str,
required=True,
help=(
"""The config json file corresponding to the pre-trained MobileBERT model. \n"""
"""This specifies the model architecture."""
),
)
parser.add_argument(
"""--pytorch_dump_path""", default=None, type=str, required=True, help="""Path to the output PyTorch model."""
)
snake_case__ : Optional[Any] = parser.parse_args()
convert_tf_checkpoint_to_pytorch(args.tf_checkpoint_path, args.mobilebert_config_file, args.pytorch_dump_path)
| 23 | 1 |
from typing import Optional, Tuple, Union
import tensorflow as tf
from ...activations_tf import ACTaFN
from ...file_utils import add_code_sample_docstrings, add_start_docstrings, add_start_docstrings_to_model_forward
from ...modeling_tf_outputs import (
TFBaseModelOutputWithNoAttention,
TFBaseModelOutputWithPoolingAndNoAttention,
TFSequenceClassifierOutput,
)
from ...modeling_tf_utils import TFPreTrainedModel, TFSequenceClassificationLoss, keras_serializable, unpack_inputs
from ...tf_utils import shape_list
from ...utils import logging
from .configuration_regnet import RegNetConfig
snake_case__ : int = logging.get_logger(__name__)
# General docstring
snake_case__ : List[str] = """RegNetConfig"""
# Base docstring
snake_case__ : Any = """facebook/regnet-y-040"""
snake_case__ : Any = [1, 1_0_8_8, 7, 7]
# Image classification docstring
snake_case__ : List[Any] = """facebook/regnet-y-040"""
snake_case__ : Union[str, Any] = """tabby, tabby cat"""
snake_case__ : Optional[int] = [
"""facebook/regnet-y-040""",
# See all regnet models at https://huggingface.co/models?filter=regnet
]
class _a ( tf.keras.layers.Layer ):
"""simple docstring"""
def __init__( self , _UpperCAmelCase , _UpperCAmelCase = 3 , _UpperCAmelCase = 1 , _UpperCAmelCase = 1 , _UpperCAmelCase = "relu" , **_UpperCAmelCase , ) -> Tuple:
super().__init__(**_UpperCAmelCase )
# The padding and conv has been verified in
# https://colab.research.google.com/gist/sayakpaul/854bc10eeaf21c9ee2119e0b9f3841a7/scratchpad.ipynb
UpperCamelCase_ = tf.keras.layers.ZeroPaddingaD(padding=kernel_size // 2 )
UpperCamelCase_ = tf.keras.layers.ConvaD(
filters=_UpperCAmelCase , kernel_size=_UpperCAmelCase , strides=_UpperCAmelCase , padding='VALID' , groups=_UpperCAmelCase , use_bias=_UpperCAmelCase , name='convolution' , )
UpperCamelCase_ = tf.keras.layers.BatchNormalization(epsilon=1e-5 , momentum=0.9 , name='normalization' )
UpperCamelCase_ = ACTaFN[activation] if activation is not None else tf.identity
def _UpperCAmelCase ( self , _UpperCAmelCase ) -> str:
UpperCamelCase_ = self.convolution(self.padding(_UpperCAmelCase ) )
UpperCamelCase_ = self.normalization(_UpperCAmelCase )
UpperCamelCase_ = self.activation(_UpperCAmelCase )
return hidden_state
class _a ( tf.keras.layers.Layer ):
"""simple docstring"""
def __init__( self , _UpperCAmelCase , **_UpperCAmelCase ) -> Any:
super().__init__(**_UpperCAmelCase )
UpperCamelCase_ = config.num_channels
UpperCamelCase_ = TFRegNetConvLayer(
out_channels=config.embedding_size , kernel_size=3 , stride=2 , activation=config.hidden_act , name='embedder' , )
def _UpperCAmelCase ( self , _UpperCAmelCase ) -> Optional[int]:
UpperCamelCase_ = shape_list(_UpperCAmelCase )[1]
if tf.executing_eagerly() and num_channels != self.num_channels:
raise ValueError(
'Make sure that the channel dimension of the pixel values match with the one set in the configuration.' )
# When running on CPU, `tf.keras.layers.Conv2D` doesn't support `NCHW` format.
# So change the input format from `NCHW` to `NHWC`.
# shape = (batch_size, in_height, in_width, in_channels=num_channels)
UpperCamelCase_ = tf.transpose(_UpperCAmelCase , perm=(0, 2, 3, 1) )
UpperCamelCase_ = self.embedder(_UpperCAmelCase )
return hidden_state
class _a ( tf.keras.layers.Layer ):
"""simple docstring"""
def __init__( self , _UpperCAmelCase , _UpperCAmelCase = 2 , **_UpperCAmelCase ) -> Union[str, Any]:
super().__init__(**_UpperCAmelCase )
UpperCamelCase_ = tf.keras.layers.ConvaD(
filters=_UpperCAmelCase , kernel_size=1 , strides=_UpperCAmelCase , use_bias=_UpperCAmelCase , name='convolution' )
UpperCamelCase_ = tf.keras.layers.BatchNormalization(epsilon=1e-5 , momentum=0.9 , name='normalization' )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase = False ) -> tf.Tensor:
return self.normalization(self.convolution(_UpperCAmelCase ) , training=_UpperCAmelCase )
class _a ( tf.keras.layers.Layer ):
"""simple docstring"""
def __init__( self , _UpperCAmelCase , _UpperCAmelCase , **_UpperCAmelCase ) -> List[str]:
super().__init__(**_UpperCAmelCase )
UpperCamelCase_ = tf.keras.layers.GlobalAveragePoolingaD(keepdims=_UpperCAmelCase , name='pooler' )
UpperCamelCase_ = [
tf.keras.layers.ConvaD(filters=_UpperCAmelCase , kernel_size=1 , activation='relu' , name='attention.0' ),
tf.keras.layers.ConvaD(filters=_UpperCAmelCase , kernel_size=1 , activation='sigmoid' , name='attention.2' ),
]
def _UpperCAmelCase ( self , _UpperCAmelCase ) -> str:
# [batch_size, h, w, num_channels] -> [batch_size, 1, 1, num_channels]
UpperCamelCase_ = self.pooler(_UpperCAmelCase )
for layer_module in self.attention:
UpperCamelCase_ = layer_module(_UpperCAmelCase )
UpperCamelCase_ = hidden_state * pooled
return hidden_state
class _a ( tf.keras.layers.Layer ):
"""simple docstring"""
def __init__( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase = 1 , **_UpperCAmelCase ) -> List[str]:
super().__init__(**_UpperCAmelCase )
UpperCamelCase_ = in_channels != out_channels or stride != 1
UpperCamelCase_ = max(1 , out_channels // config.groups_width )
UpperCamelCase_ = (
TFRegNetShortCut(_UpperCAmelCase , stride=_UpperCAmelCase , name='shortcut' )
if should_apply_shortcut
else tf.keras.layers.Activation('linear' , name='shortcut' )
)
# `self.layers` instead of `self.layer` because that is a reserved argument.
UpperCamelCase_ = [
TFRegNetConvLayer(_UpperCAmelCase , kernel_size=1 , activation=config.hidden_act , name='layer.0' ),
TFRegNetConvLayer(
_UpperCAmelCase , stride=_UpperCAmelCase , groups=_UpperCAmelCase , activation=config.hidden_act , name='layer.1' ),
TFRegNetConvLayer(_UpperCAmelCase , kernel_size=1 , activation=_UpperCAmelCase , name='layer.2' ),
]
UpperCamelCase_ = ACTaFN[config.hidden_act]
def _UpperCAmelCase ( self , _UpperCAmelCase ) -> Tuple:
UpperCamelCase_ = hidden_state
for layer_module in self.layers:
UpperCamelCase_ = layer_module(_UpperCAmelCase )
UpperCamelCase_ = self.shortcut(_UpperCAmelCase )
hidden_state += residual
UpperCamelCase_ = self.activation(_UpperCAmelCase )
return hidden_state
class _a ( tf.keras.layers.Layer ):
"""simple docstring"""
def __init__( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase = 1 , **_UpperCAmelCase ) -> Optional[int]:
super().__init__(**_UpperCAmelCase )
UpperCamelCase_ = in_channels != out_channels or stride != 1
UpperCamelCase_ = max(1 , out_channels // config.groups_width )
UpperCamelCase_ = (
TFRegNetShortCut(_UpperCAmelCase , stride=_UpperCAmelCase , name='shortcut' )
if should_apply_shortcut
else tf.keras.layers.Activation('linear' , name='shortcut' )
)
UpperCamelCase_ = [
TFRegNetConvLayer(_UpperCAmelCase , kernel_size=1 , activation=config.hidden_act , name='layer.0' ),
TFRegNetConvLayer(
_UpperCAmelCase , stride=_UpperCAmelCase , groups=_UpperCAmelCase , activation=config.hidden_act , name='layer.1' ),
TFRegNetSELayer(_UpperCAmelCase , reduced_channels=int(round(in_channels / 4 ) ) , name='layer.2' ),
TFRegNetConvLayer(_UpperCAmelCase , kernel_size=1 , activation=_UpperCAmelCase , name='layer.3' ),
]
UpperCamelCase_ = ACTaFN[config.hidden_act]
def _UpperCAmelCase ( self , _UpperCAmelCase ) -> Tuple:
UpperCamelCase_ = hidden_state
for layer_module in self.layers:
UpperCamelCase_ = layer_module(_UpperCAmelCase )
UpperCamelCase_ = self.shortcut(_UpperCAmelCase )
hidden_state += residual
UpperCamelCase_ = self.activation(_UpperCAmelCase )
return hidden_state
class _a ( tf.keras.layers.Layer ):
"""simple docstring"""
def __init__( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase = 2 , _UpperCAmelCase = 2 , **_UpperCAmelCase ) -> List[Any]:
super().__init__(**_UpperCAmelCase )
UpperCamelCase_ = TFRegNetXLayer if config.layer_type == 'x' else TFRegNetYLayer
UpperCamelCase_ = [
# downsampling is done in the first layer with stride of 2
layer(_UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , stride=_UpperCAmelCase , name='layers.0' ),
*[layer(_UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , name=f"""layers.{i+1}""" ) for i in range(depth - 1 )],
]
def _UpperCAmelCase ( self , _UpperCAmelCase ) -> Union[str, Any]:
for layer_module in self.layers:
UpperCamelCase_ = layer_module(_UpperCAmelCase )
return hidden_state
class _a ( tf.keras.layers.Layer ):
"""simple docstring"""
def __init__( self , _UpperCAmelCase , **_UpperCAmelCase ) -> str:
super().__init__(**_UpperCAmelCase )
UpperCamelCase_ = []
# based on `downsample_in_first_stage`, the first layer of the first stage may or may not downsample the input
self.stages.append(
TFRegNetStage(
_UpperCAmelCase , config.embedding_size , config.hidden_sizes[0] , stride=2 if config.downsample_in_first_stage else 1 , depth=config.depths[0] , name='stages.0' , ) )
UpperCamelCase_ = zip(config.hidden_sizes , config.hidden_sizes[1:] )
for i, ((in_channels, out_channels), depth) in enumerate(zip(_UpperCAmelCase , config.depths[1:] ) ):
self.stages.append(TFRegNetStage(_UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , depth=_UpperCAmelCase , name=f"""stages.{i+1}""" ) )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase = False , _UpperCAmelCase = True ) -> TFBaseModelOutputWithNoAttention:
UpperCamelCase_ = () if output_hidden_states else None
for stage_module in self.stages:
if output_hidden_states:
UpperCamelCase_ = hidden_states + (hidden_state,)
UpperCamelCase_ = stage_module(_UpperCAmelCase )
if output_hidden_states:
UpperCamelCase_ = hidden_states + (hidden_state,)
if not return_dict:
return tuple(v for v in [hidden_state, hidden_states] if v is not None )
return TFBaseModelOutputWithNoAttention(last_hidden_state=_UpperCAmelCase , hidden_states=_UpperCAmelCase )
@keras_serializable
class _a ( tf.keras.layers.Layer ):
"""simple docstring"""
A_ = RegNetConfig
def __init__( self , _UpperCAmelCase , **_UpperCAmelCase ) -> Tuple:
super().__init__(**_UpperCAmelCase )
UpperCamelCase_ = config
UpperCamelCase_ = TFRegNetEmbeddings(_UpperCAmelCase , name='embedder' )
UpperCamelCase_ = TFRegNetEncoder(_UpperCAmelCase , name='encoder' )
UpperCamelCase_ = tf.keras.layers.GlobalAveragePoolingaD(keepdims=_UpperCAmelCase , name='pooler' )
@unpack_inputs
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = False , ) -> TFBaseModelOutputWithPoolingAndNoAttention:
UpperCamelCase_ = (
output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states
)
UpperCamelCase_ = return_dict if return_dict is not None else self.config.use_return_dict
UpperCamelCase_ = self.embedder(_UpperCAmelCase , training=_UpperCAmelCase )
UpperCamelCase_ = self.encoder(
_UpperCAmelCase , output_hidden_states=_UpperCAmelCase , return_dict=_UpperCAmelCase , training=_UpperCAmelCase )
UpperCamelCase_ = encoder_outputs[0]
UpperCamelCase_ = self.pooler(_UpperCAmelCase )
# Change to NCHW output format have uniformity in the modules
UpperCamelCase_ = tf.transpose(_UpperCAmelCase , perm=(0, 3, 1, 2) )
UpperCamelCase_ = tf.transpose(_UpperCAmelCase , perm=(0, 3, 1, 2) )
# Change the other hidden state outputs to NCHW as well
if output_hidden_states:
UpperCamelCase_ = tuple([tf.transpose(_UpperCAmelCase , perm=(0, 3, 1, 2) ) for h in encoder_outputs[1]] )
if not return_dict:
return (last_hidden_state, pooled_output) + encoder_outputs[1:]
return TFBaseModelOutputWithPoolingAndNoAttention(
last_hidden_state=_UpperCAmelCase , pooler_output=_UpperCAmelCase , hidden_states=hidden_states if output_hidden_states else encoder_outputs.hidden_states , )
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
A_ = RegNetConfig
A_ = """regnet"""
A_ = """pixel_values"""
@property
def _UpperCAmelCase ( self ) -> Tuple:
return {"pixel_values": tf.TensorSpec(shape=(None, self.config.num_channels, 224, 224) , dtype=tf.floataa )}
snake_case__ : Dict = R"""
Parameters:
This model is a Tensorflow
[tf.keras.layers.Layer](https://www.tensorflow.org/api_docs/python/tf/keras/layers/Layer) sub-class. Use it as a
regular Tensorflow Module and refer to the Tensorflow documentation for all matter related to general usage and
behavior.
config ([`RegNetConfig`]): Model configuration class with all the parameters of the model.
Initializing with a config file does not load the weights associated with the model, only the
configuration. Check out the [`~TFPreTrainedModel.from_pretrained`] method to load the model weights.
"""
snake_case__ : Optional[Any] = R"""
Args:
pixel_values (`tf.Tensor` of shape `(batch_size, num_channels, height, width)`):
Pixel values. Pixel values can be obtained using [`AutoImageProcessor`]. See
[`ConveNextImageProcessor.__call__`] for details.
output_hidden_states (`bool`, *optional*):
Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors for
more detail.
return_dict (`bool`, *optional*):
Whether or not to return a [`~utils.ModelOutput`] instead of a plain tuple.
"""
@add_start_docstrings(
"""The bare RegNet model outputting raw features without any specific head on top.""" , UpperCAmelCase__ , )
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
def __init__( self , _UpperCAmelCase , *_UpperCAmelCase , **_UpperCAmelCase ) -> List[Any]:
super().__init__(_UpperCAmelCase , *_UpperCAmelCase , **_UpperCAmelCase )
UpperCamelCase_ = TFRegNetMainLayer(_UpperCAmelCase , name='regnet' )
@unpack_inputs
@add_start_docstrings_to_model_forward(_UpperCAmelCase )
@add_code_sample_docstrings(
checkpoint=_CHECKPOINT_FOR_DOC , output_type=_UpperCAmelCase , config_class=_CONFIG_FOR_DOC , modality='vision' , expected_output=_EXPECTED_OUTPUT_SHAPE , )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase=False , ) -> Union[TFBaseModelOutputWithPoolingAndNoAttention, Tuple[tf.Tensor]]:
UpperCamelCase_ = (
output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states
)
UpperCamelCase_ = return_dict if return_dict is not None else self.config.use_return_dict
UpperCamelCase_ = self.regnet(
pixel_values=_UpperCAmelCase , output_hidden_states=_UpperCAmelCase , return_dict=_UpperCAmelCase , training=_UpperCAmelCase , )
if not return_dict:
return (outputs[0],) + outputs[1:]
return TFBaseModelOutputWithPoolingAndNoAttention(
last_hidden_state=outputs.last_hidden_state , pooler_output=outputs.pooler_output , hidden_states=outputs.hidden_states , )
@add_start_docstrings(
"""
RegNet Model with an image classification head on top (a linear layer on top of the pooled features), e.g. for
ImageNet.
""" , UpperCAmelCase__ , )
class _a ( UpperCAmelCase__ , UpperCAmelCase__ ):
"""simple docstring"""
def __init__( self , _UpperCAmelCase , *_UpperCAmelCase , **_UpperCAmelCase ) -> List[Any]:
super().__init__(_UpperCAmelCase , *_UpperCAmelCase , **_UpperCAmelCase )
UpperCamelCase_ = config.num_labels
UpperCamelCase_ = TFRegNetMainLayer(_UpperCAmelCase , name='regnet' )
# classification head
UpperCamelCase_ = [
tf.keras.layers.Flatten(),
tf.keras.layers.Dense(config.num_labels , name='classifier.1' ) if config.num_labels > 0 else tf.identity,
]
@unpack_inputs
@add_start_docstrings_to_model_forward(_UpperCAmelCase )
@add_code_sample_docstrings(
checkpoint=_IMAGE_CLASS_CHECKPOINT , output_type=_UpperCAmelCase , config_class=_CONFIG_FOR_DOC , expected_output=_IMAGE_CLASS_EXPECTED_OUTPUT , )
def _UpperCAmelCase ( self , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase=False , ) -> Union[TFSequenceClassifierOutput, Tuple[tf.Tensor]]:
UpperCamelCase_ = (
output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states
)
UpperCamelCase_ = return_dict if return_dict is not None else self.config.use_return_dict
UpperCamelCase_ = self.regnet(
_UpperCAmelCase , output_hidden_states=_UpperCAmelCase , return_dict=_UpperCAmelCase , training=_UpperCAmelCase )
UpperCamelCase_ = outputs.pooler_output if return_dict else outputs[1]
UpperCamelCase_ = self.classifier[0](_UpperCAmelCase )
UpperCamelCase_ = self.classifier[1](_UpperCAmelCase )
UpperCamelCase_ = None if labels is None else self.hf_compute_loss(labels=_UpperCAmelCase , logits=_UpperCAmelCase )
if not return_dict:
UpperCamelCase_ = (logits,) + outputs[2:]
return ((loss,) + output) if loss is not None else output
return TFSequenceClassifierOutput(loss=_UpperCAmelCase , logits=_UpperCAmelCase , hidden_states=outputs.hidden_states )
| 23 |
import gc
import unittest
from transformers import MODEL_FOR_MASKED_LM_MAPPING, TF_MODEL_FOR_MASKED_LM_MAPPING, FillMaskPipeline, pipeline
from transformers.pipelines import PipelineException
from transformers.testing_utils import (
is_pipeline_test,
is_torch_available,
nested_simplify,
require_tf,
require_torch,
require_torch_gpu,
slow,
)
from .test_pipelines_common import ANY
@is_pipeline_test
class _a ( unittest.TestCase ):
"""simple docstring"""
A_ = MODEL_FOR_MASKED_LM_MAPPING
A_ = TF_MODEL_FOR_MASKED_LM_MAPPING
def _UpperCAmelCase ( self ) -> List[str]:
super().tearDown()
# clean-up as much as possible GPU memory occupied by PyTorch
gc.collect()
if is_torch_available():
import torch
torch.cuda.empty_cache()
@require_tf
def _UpperCAmelCase ( self ) -> str:
UpperCamelCase_ = pipeline(task='fill-mask' , model='sshleifer/tiny-distilroberta-base' , top_k=2 , framework='tf' )
UpperCamelCase_ = unmasker('My name is <mask>' )
self.assertEqual(
nested_simplify(_UpperCAmelCase , decimals=6 ) , [
{'sequence': 'My name is grouped', 'score': 2.1e-05, 'token': 38015, 'token_str': ' grouped'},
{'sequence': 'My name is accuser', 'score': 2.1e-05, 'token': 25506, 'token_str': ' accuser'},
] , )
UpperCamelCase_ = unmasker('The largest city in France is <mask>' )
self.assertEqual(
nested_simplify(_UpperCAmelCase , decimals=6 ) , [
{
'sequence': 'The largest city in France is grouped',
'score': 2.1e-05,
'token': 38015,
'token_str': ' grouped',
},
{
'sequence': 'The largest city in France is accuser',
'score': 2.1e-05,
'token': 25506,
'token_str': ' accuser',
},
] , )
UpperCamelCase_ = unmasker('My name is <mask>' , targets=[' Patrick', ' Clara', ' Teven'] , top_k=3 )
self.assertEqual(
nested_simplify(_UpperCAmelCase , decimals=6 ) , [
{'sequence': 'My name is Clara', 'score': 2e-05, 'token': 13606, 'token_str': ' Clara'},
{'sequence': 'My name is Patrick', 'score': 2e-05, 'token': 3499, 'token_str': ' Patrick'},
{'sequence': 'My name is Te', 'score': 1.9e-05, 'token': 2941, 'token_str': ' Te'},
] , )
@require_torch
def _UpperCAmelCase ( self ) -> Union[str, Any]:
UpperCamelCase_ = pipeline(task='fill-mask' , model='sshleifer/tiny-distilroberta-base' , top_k=2 , framework='pt' )
UpperCamelCase_ = unmasker('My name is <mask>' )
self.assertEqual(
nested_simplify(_UpperCAmelCase , decimals=6 ) , [
{'sequence': 'My name is Maul', 'score': 2.2e-05, 'token': 35676, 'token_str': ' Maul'},
{'sequence': 'My name isELS', 'score': 2.2e-05, 'token': 16416, 'token_str': 'ELS'},
] , )
UpperCamelCase_ = unmasker('The largest city in France is <mask>' )
self.assertEqual(
nested_simplify(_UpperCAmelCase , decimals=6 ) , [
{
'sequence': 'The largest city in France is Maul',
'score': 2.2e-05,
'token': 35676,
'token_str': ' Maul',
},
{'sequence': 'The largest city in France isELS', 'score': 2.2e-05, 'token': 16416, 'token_str': 'ELS'},
] , )
UpperCamelCase_ = unmasker('My name is <mask>' , targets=[' Patrick', ' Clara', ' Teven'] , top_k=3 )
self.assertEqual(
nested_simplify(_UpperCAmelCase , decimals=6 ) , [
{'sequence': 'My name is Patrick', 'score': 2.1e-05, 'token': 3499, 'token_str': ' Patrick'},
{'sequence': 'My name is Te', 'score': 2e-05, 'token': 2941, 'token_str': ' Te'},
{'sequence': 'My name is Clara', 'score': 2e-05, 'token': 13606, 'token_str': ' Clara'},
] , )
UpperCamelCase_ = unmasker('My name is <mask> <mask>' , top_k=2 )
self.assertEqual(
nested_simplify(_UpperCAmelCase , decimals=6 ) , [
[
{
'score': 2.2e-05,
'token': 35676,
'token_str': ' Maul',
'sequence': '<s>My name is Maul<mask></s>',
},
{'score': 2.2e-05, 'token': 16416, 'token_str': 'ELS', 'sequence': '<s>My name isELS<mask></s>'},
],
[
{
'score': 2.2e-05,
'token': 35676,
'token_str': ' Maul',
'sequence': '<s>My name is<mask> Maul</s>',
},
{'score': 2.2e-05, 'token': 16416, 'token_str': 'ELS', 'sequence': '<s>My name is<mask>ELS</s>'},
],
] , )
@require_torch_gpu
def _UpperCAmelCase ( self ) -> Optional[Any]:
UpperCamelCase_ = pipeline('fill-mask' , model='hf-internal-testing/tiny-random-distilbert' , device=0 , framework='pt' )
# convert model to fp16
pipe.model.half()
UpperCamelCase_ = pipe('Paris is the [MASK] of France.' )
# We actually don't care about the result, we just want to make sure
# it works, meaning the float16 tensor got casted back to float32
# for postprocessing.
self.assertIsInstance(_UpperCAmelCase , _UpperCAmelCase )
@slow
@require_torch
def _UpperCAmelCase ( self ) -> List[Any]:
UpperCamelCase_ = pipeline(task='fill-mask' , model='distilroberta-base' , top_k=2 , framework='pt' )
self.run_large_test(_UpperCAmelCase )
@slow
@require_tf
def _UpperCAmelCase ( self ) -> Union[str, Any]:
UpperCamelCase_ = pipeline(task='fill-mask' , model='distilroberta-base' , top_k=2 , framework='tf' )
self.run_large_test(_UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase ) -> Tuple:
UpperCamelCase_ = unmasker('My name is <mask>' )
self.assertEqual(
nested_simplify(_UpperCAmelCase ) , [
{'sequence': 'My name is John', 'score': 0.0_0_8, 'token': 610, 'token_str': ' John'},
{'sequence': 'My name is Chris', 'score': 0.0_0_7, 'token': 1573, 'token_str': ' Chris'},
] , )
UpperCamelCase_ = unmasker('The largest city in France is <mask>' )
self.assertEqual(
nested_simplify(_UpperCAmelCase ) , [
{
'sequence': 'The largest city in France is Paris',
'score': 0.2_5_1,
'token': 2201,
'token_str': ' Paris',
},
{
'sequence': 'The largest city in France is Lyon',
'score': 0.2_1_4,
'token': 12790,
'token_str': ' Lyon',
},
] , )
UpperCamelCase_ = unmasker('My name is <mask>' , targets=[' Patrick', ' Clara', ' Teven'] , top_k=3 )
self.assertEqual(
nested_simplify(_UpperCAmelCase ) , [
{'sequence': 'My name is Patrick', 'score': 0.0_0_5, 'token': 3499, 'token_str': ' Patrick'},
{'sequence': 'My name is Clara', 'score': 0.0_0_0, 'token': 13606, 'token_str': ' Clara'},
{'sequence': 'My name is Te', 'score': 0.0_0_0, 'token': 2941, 'token_str': ' Te'},
] , )
@require_torch
def _UpperCAmelCase ( self ) -> List[Any]:
UpperCamelCase_ = pipeline(task='fill-mask' , model='sshleifer/tiny-distilroberta-base' , framework='pt' )
UpperCamelCase_ = None
UpperCamelCase_ = None
self.run_pipeline_test(_UpperCAmelCase , [] )
@require_tf
def _UpperCAmelCase ( self ) -> Optional[int]:
UpperCamelCase_ = pipeline(task='fill-mask' , model='sshleifer/tiny-distilroberta-base' , framework='tf' )
UpperCamelCase_ = None
UpperCamelCase_ = None
self.run_pipeline_test(_UpperCAmelCase , [] )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase ) -> Optional[Any]:
if tokenizer is None or tokenizer.mask_token_id is None:
self.skipTest('The provided tokenizer has no mask token, (probably reformer or wav2vec2)' )
UpperCamelCase_ = FillMaskPipeline(model=_UpperCAmelCase , tokenizer=_UpperCAmelCase )
UpperCamelCase_ = [
f"""This is another {tokenizer.mask_token} test""",
]
return fill_masker, examples
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> Union[str, Any]:
UpperCamelCase_ = fill_masker.tokenizer
UpperCamelCase_ = fill_masker.model
UpperCamelCase_ = fill_masker(
f"""This is a {tokenizer.mask_token}""" , )
self.assertEqual(
_UpperCAmelCase , [
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
] , )
UpperCamelCase_ = fill_masker([f"""This is a {tokenizer.mask_token}"""] )
self.assertEqual(
_UpperCAmelCase , [
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
] , )
UpperCamelCase_ = fill_masker([f"""This is a {tokenizer.mask_token}""", f"""Another {tokenizer.mask_token} great test."""] )
self.assertEqual(
_UpperCAmelCase , [
[
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
],
[
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
],
] , )
with self.assertRaises(_UpperCAmelCase ):
fill_masker([None] )
# No mask_token is not supported
with self.assertRaises(_UpperCAmelCase ):
fill_masker('This is' )
self.run_test_top_k(_UpperCAmelCase , _UpperCAmelCase )
self.run_test_targets(_UpperCAmelCase , _UpperCAmelCase )
self.run_test_top_k_targets(_UpperCAmelCase , _UpperCAmelCase )
self.fill_mask_with_duplicate_targets_and_top_k(_UpperCAmelCase , _UpperCAmelCase )
self.fill_mask_with_multiple_masks(_UpperCAmelCase , _UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> Optional[Any]:
UpperCamelCase_ = tokenizer.get_vocab()
UpperCamelCase_ = sorted(vocab.keys() )[:2]
# Pipeline argument
UpperCamelCase_ = FillMaskPipeline(model=_UpperCAmelCase , tokenizer=_UpperCAmelCase , targets=_UpperCAmelCase )
UpperCamelCase_ = fill_masker(f"""This is a {tokenizer.mask_token}""" )
self.assertEqual(
_UpperCAmelCase , [
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
] , )
UpperCamelCase_ = {vocab[el] for el in targets}
self.assertEqual({el['token'] for el in outputs} , _UpperCAmelCase )
UpperCamelCase_ = [tokenizer.decode([x] ) for x in target_ids]
self.assertEqual({el['token_str'] for el in outputs} , set(_UpperCAmelCase ) )
# Call argument
UpperCamelCase_ = FillMaskPipeline(model=_UpperCAmelCase , tokenizer=_UpperCAmelCase )
UpperCamelCase_ = fill_masker(f"""This is a {tokenizer.mask_token}""" , targets=_UpperCAmelCase )
self.assertEqual(
_UpperCAmelCase , [
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
] , )
UpperCamelCase_ = {vocab[el] for el in targets}
self.assertEqual({el['token'] for el in outputs} , _UpperCAmelCase )
UpperCamelCase_ = [tokenizer.decode([x] ) for x in target_ids]
self.assertEqual({el['token_str'] for el in outputs} , set(_UpperCAmelCase ) )
# Score equivalence
UpperCamelCase_ = fill_masker(f"""This is a {tokenizer.mask_token}""" , targets=_UpperCAmelCase )
UpperCamelCase_ = [top_mask['token_str'] for top_mask in outputs]
UpperCamelCase_ = [top_mask['score'] for top_mask in outputs]
# For some BPE tokenizers, `</w>` is removed during decoding, so `token_str` won't be the same as in `targets`.
if set(_UpperCAmelCase ) == set(_UpperCAmelCase ):
UpperCamelCase_ = fill_masker(f"""This is a {tokenizer.mask_token}""" , targets=_UpperCAmelCase )
UpperCamelCase_ = [top_mask['score'] for top_mask in unmasked_targets]
self.assertEqual(nested_simplify(_UpperCAmelCase ) , nested_simplify(_UpperCAmelCase ) )
# Raises with invalid
with self.assertRaises(_UpperCAmelCase ):
UpperCamelCase_ = fill_masker(f"""This is a {tokenizer.mask_token}""" , targets=[] )
# For some tokenizers, `""` is actually in the vocabulary and the expected error won't raised
if "" not in tokenizer.get_vocab():
with self.assertRaises(_UpperCAmelCase ):
UpperCamelCase_ = fill_masker(f"""This is a {tokenizer.mask_token}""" , targets=[''] )
with self.assertRaises(_UpperCAmelCase ):
UpperCamelCase_ = fill_masker(f"""This is a {tokenizer.mask_token}""" , targets='' )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> Tuple:
UpperCamelCase_ = FillMaskPipeline(model=_UpperCAmelCase , tokenizer=_UpperCAmelCase , top_k=2 )
UpperCamelCase_ = fill_masker(f"""This is a {tokenizer.mask_token}""" )
self.assertEqual(
_UpperCAmelCase , [
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
] , )
UpperCamelCase_ = FillMaskPipeline(model=_UpperCAmelCase , tokenizer=_UpperCAmelCase )
UpperCamelCase_ = fill_masker(f"""This is a {tokenizer.mask_token}""" , top_k=2 )
self.assertEqual(
_UpperCAmelCase , [
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
] , )
self.assertEqual(nested_simplify(_UpperCAmelCase ) , nested_simplify(_UpperCAmelCase ) )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> List[Any]:
UpperCamelCase_ = tokenizer.get_vocab()
UpperCamelCase_ = FillMaskPipeline(model=_UpperCAmelCase , tokenizer=_UpperCAmelCase )
# top_k=2, ntargets=3
UpperCamelCase_ = sorted(vocab.keys() )[:3]
UpperCamelCase_ = fill_masker(f"""This is a {tokenizer.mask_token}""" , top_k=2 , targets=_UpperCAmelCase )
# If we use the most probably targets, and filter differently, we should still
# have the same results
UpperCamelCase_ = [el['token_str'] for el in sorted(_UpperCAmelCase , key=lambda _UpperCAmelCase : x["score"] , reverse=_UpperCAmelCase )]
# For some BPE tokenizers, `</w>` is removed during decoding, so `token_str` won't be the same as in `targets`.
if set(_UpperCAmelCase ).issubset(_UpperCAmelCase ):
UpperCamelCase_ = fill_masker(f"""This is a {tokenizer.mask_token}""" , top_k=3 , targets=_UpperCAmelCase )
# They should yield exactly the same result
self.assertEqual(nested_simplify(_UpperCAmelCase ) , nested_simplify(_UpperCAmelCase ) )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> Optional[int]:
UpperCamelCase_ = FillMaskPipeline(model=_UpperCAmelCase , tokenizer=_UpperCAmelCase )
UpperCamelCase_ = tokenizer.get_vocab()
# String duplicates + id duplicates
UpperCamelCase_ = sorted(vocab.keys() )[:3]
UpperCamelCase_ = [targets[0], targets[1], targets[0], targets[2], targets[1]]
UpperCamelCase_ = fill_masker(f"""My name is {tokenizer.mask_token}""" , targets=_UpperCAmelCase , top_k=10 )
# The target list contains duplicates, so we can't output more
# than them
self.assertEqual(len(_UpperCAmelCase ) , 3 )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> List[str]:
UpperCamelCase_ = FillMaskPipeline(model=_UpperCAmelCase , tokenizer=_UpperCAmelCase )
UpperCamelCase_ = fill_masker(
f"""This is a {tokenizer.mask_token} {tokenizer.mask_token} {tokenizer.mask_token}""" , top_k=2 )
self.assertEqual(
_UpperCAmelCase , [
[
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
],
[
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
],
[
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
],
] , )
| 23 | 1 |
import numpy as np
from matplotlib import pyplot as plt
from sklearn.datasets import load_iris
from sklearn.metrics import ConfusionMatrixDisplay
from sklearn.model_selection import train_test_split
from xgboost import XGBClassifier
def _snake_case (__lowercase):
return (data["data"], data["target"])
def _snake_case (__lowercase , __lowercase):
UpperCamelCase_ = XGBClassifier()
classifier.fit(__lowercase , __lowercase)
return classifier
def _snake_case ():
UpperCamelCase_ = load_iris()
UpperCamelCase_ , UpperCamelCase_ = data_handling(__lowercase)
UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ = train_test_split(
__lowercase , __lowercase , test_size=0.25)
UpperCamelCase_ = iris['target_names']
# Create an XGBoost Classifier from the training data
UpperCamelCase_ = xgboost(__lowercase , __lowercase)
# Display the confusion matrix of the classifier with both training and test sets
ConfusionMatrixDisplay.from_estimator(
__lowercase , __lowercase , __lowercase , display_labels=__lowercase , cmap='Blues' , normalize='true' , )
plt.title('Normalized Confusion Matrix - IRIS Dataset')
plt.show()
if __name__ == "__main__":
import doctest
doctest.testmod(verbose=True)
main()
| 23 |
import gc
import unittest
import numpy as np
import torch
from transformers import CLIPTextConfig, CLIPTextModel, CLIPTokenizer
from diffusers import (
AutoencoderKL,
DDIMScheduler,
StableDiffusionSAGPipeline,
UNetaDConditionModel,
)
from diffusers.utils import slow, torch_device
from diffusers.utils.testing_utils import enable_full_determinism, require_torch_gpu
from ..pipeline_params import TEXT_TO_IMAGE_BATCH_PARAMS, TEXT_TO_IMAGE_IMAGE_PARAMS, TEXT_TO_IMAGE_PARAMS
from ..test_pipelines_common import PipelineLatentTesterMixin, PipelineTesterMixin
enable_full_determinism()
class _a ( UpperCAmelCase__ , UpperCAmelCase__ , unittest.TestCase ):
"""simple docstring"""
A_ = StableDiffusionSAGPipeline
A_ = TEXT_TO_IMAGE_PARAMS
A_ = TEXT_TO_IMAGE_BATCH_PARAMS
A_ = TEXT_TO_IMAGE_IMAGE_PARAMS
A_ = TEXT_TO_IMAGE_IMAGE_PARAMS
A_ = False
def _UpperCAmelCase ( self ) -> Optional[Any]:
torch.manual_seed(0 )
UpperCamelCase_ = UNetaDConditionModel(
block_out_channels=(32, 64) , layers_per_block=2 , sample_size=32 , in_channels=4 , out_channels=4 , down_block_types=('DownBlock2D', 'CrossAttnDownBlock2D') , up_block_types=('CrossAttnUpBlock2D', 'UpBlock2D') , cross_attention_dim=32 , )
UpperCamelCase_ = DDIMScheduler(
beta_start=0.0_0_0_8_5 , beta_end=0.0_1_2 , beta_schedule='scaled_linear' , clip_sample=_UpperCAmelCase , set_alpha_to_one=_UpperCAmelCase , )
torch.manual_seed(0 )
UpperCamelCase_ = AutoencoderKL(
block_out_channels=[32, 64] , in_channels=3 , out_channels=3 , down_block_types=['DownEncoderBlock2D', 'DownEncoderBlock2D'] , up_block_types=['UpDecoderBlock2D', 'UpDecoderBlock2D'] , latent_channels=4 , )
torch.manual_seed(0 )
UpperCamelCase_ = CLIPTextConfig(
bos_token_id=0 , eos_token_id=2 , hidden_size=32 , intermediate_size=37 , layer_norm_eps=1e-05 , num_attention_heads=4 , num_hidden_layers=5 , pad_token_id=1 , vocab_size=1000 , )
UpperCamelCase_ = CLIPTextModel(_UpperCAmelCase )
UpperCamelCase_ = CLIPTokenizer.from_pretrained('hf-internal-testing/tiny-random-clip' )
UpperCamelCase_ = {
'unet': unet,
'scheduler': scheduler,
'vae': vae,
'text_encoder': text_encoder,
'tokenizer': tokenizer,
'safety_checker': None,
'feature_extractor': None,
}
return components
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase=0 ) -> List[Any]:
if str(_UpperCAmelCase ).startswith('mps' ):
UpperCamelCase_ = torch.manual_seed(_UpperCAmelCase )
else:
UpperCamelCase_ = torch.Generator(device=_UpperCAmelCase ).manual_seed(_UpperCAmelCase )
UpperCamelCase_ = {
'prompt': '.',
'generator': generator,
'num_inference_steps': 2,
'guidance_scale': 1.0,
'sag_scale': 1.0,
'output_type': 'numpy',
}
return inputs
def _UpperCAmelCase ( self ) -> Tuple:
super().test_inference_batch_single_identical(expected_max_diff=3e-3 )
@slow
@require_torch_gpu
class _a ( unittest.TestCase ):
"""simple docstring"""
def _UpperCAmelCase ( self ) -> Tuple:
# clean up the VRAM after each test
super().tearDown()
gc.collect()
torch.cuda.empty_cache()
def _UpperCAmelCase ( self ) -> str:
UpperCamelCase_ = StableDiffusionSAGPipeline.from_pretrained('CompVis/stable-diffusion-v1-4' )
UpperCamelCase_ = sag_pipe.to(_UpperCAmelCase )
sag_pipe.set_progress_bar_config(disable=_UpperCAmelCase )
UpperCamelCase_ = '.'
UpperCamelCase_ = torch.manual_seed(0 )
UpperCamelCase_ = sag_pipe(
[prompt] , generator=_UpperCAmelCase , guidance_scale=7.5 , sag_scale=1.0 , num_inference_steps=20 , output_type='np' )
UpperCamelCase_ = output.images
UpperCamelCase_ = image[0, -3:, -3:, -1]
assert image.shape == (1, 512, 512, 3)
UpperCamelCase_ = np.array([0.1_5_6_8, 0.1_7_3_8, 0.1_6_9_5, 0.1_6_9_3, 0.1_5_0_7, 0.1_7_0_5, 0.1_5_4_7, 0.1_7_5_1, 0.1_9_4_9] )
assert np.abs(image_slice.flatten() - expected_slice ).max() < 5e-2
def _UpperCAmelCase ( self ) -> List[Any]:
UpperCamelCase_ = StableDiffusionSAGPipeline.from_pretrained('stabilityai/stable-diffusion-2-1-base' )
UpperCamelCase_ = sag_pipe.to(_UpperCAmelCase )
sag_pipe.set_progress_bar_config(disable=_UpperCAmelCase )
UpperCamelCase_ = '.'
UpperCamelCase_ = torch.manual_seed(0 )
UpperCamelCase_ = sag_pipe(
[prompt] , generator=_UpperCAmelCase , guidance_scale=7.5 , sag_scale=1.0 , num_inference_steps=20 , output_type='np' )
UpperCamelCase_ = output.images
UpperCamelCase_ = image[0, -3:, -3:, -1]
assert image.shape == (1, 512, 512, 3)
UpperCamelCase_ = np.array([0.3_4_5_9, 0.2_8_7_6, 0.2_5_3_7, 0.3_0_0_2, 0.2_6_7_1, 0.2_1_6_0, 0.3_0_2_6, 0.2_2_6_2, 0.2_3_7_1] )
assert np.abs(image_slice.flatten() - expected_slice ).max() < 5e-2
def _UpperCAmelCase ( self ) -> Dict:
UpperCamelCase_ = StableDiffusionSAGPipeline.from_pretrained('stabilityai/stable-diffusion-2-1-base' )
UpperCamelCase_ = sag_pipe.to(_UpperCAmelCase )
sag_pipe.set_progress_bar_config(disable=_UpperCAmelCase )
UpperCamelCase_ = '.'
UpperCamelCase_ = torch.manual_seed(0 )
UpperCamelCase_ = sag_pipe(
[prompt] , width=768 , height=512 , generator=_UpperCAmelCase , guidance_scale=7.5 , sag_scale=1.0 , num_inference_steps=20 , output_type='np' , )
UpperCamelCase_ = output.images
assert image.shape == (1, 512, 768, 3)
| 23 | 1 |
from typing import TYPE_CHECKING
from ...utils import (
OptionalDependencyNotAvailable,
_LazyModule,
is_sentencepiece_available,
is_tf_available,
is_tokenizers_available,
is_torch_available,
)
snake_case__ : Dict = {"""configuration_xlnet""": ["""XLNET_PRETRAINED_CONFIG_ARCHIVE_MAP""", """XLNetConfig"""]}
try:
if not is_sentencepiece_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
snake_case__ : Dict = ["""XLNetTokenizer"""]
try:
if not is_tokenizers_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
snake_case__ : int = ["""XLNetTokenizerFast"""]
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
snake_case__ : Any = [
"""XLNET_PRETRAINED_MODEL_ARCHIVE_LIST""",
"""XLNetForMultipleChoice""",
"""XLNetForQuestionAnswering""",
"""XLNetForQuestionAnsweringSimple""",
"""XLNetForSequenceClassification""",
"""XLNetForTokenClassification""",
"""XLNetLMHeadModel""",
"""XLNetModel""",
"""XLNetPreTrainedModel""",
"""load_tf_weights_in_xlnet""",
]
try:
if not is_tf_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
snake_case__ : Optional[Any] = [
"""TF_XLNET_PRETRAINED_MODEL_ARCHIVE_LIST""",
"""TFXLNetForMultipleChoice""",
"""TFXLNetForQuestionAnsweringSimple""",
"""TFXLNetForSequenceClassification""",
"""TFXLNetForTokenClassification""",
"""TFXLNetLMHeadModel""",
"""TFXLNetMainLayer""",
"""TFXLNetModel""",
"""TFXLNetPreTrainedModel""",
]
if TYPE_CHECKING:
from .configuration_xlnet import XLNET_PRETRAINED_CONFIG_ARCHIVE_MAP, XLNetConfig
try:
if not is_sentencepiece_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .tokenization_xlnet import XLNetTokenizer
try:
if not is_tokenizers_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .tokenization_xlnet_fast import XLNetTokenizerFast
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_xlnet import (
XLNET_PRETRAINED_MODEL_ARCHIVE_LIST,
XLNetForMultipleChoice,
XLNetForQuestionAnswering,
XLNetForQuestionAnsweringSimple,
XLNetForSequenceClassification,
XLNetForTokenClassification,
XLNetLMHeadModel,
XLNetModel,
XLNetPreTrainedModel,
load_tf_weights_in_xlnet,
)
try:
if not is_tf_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_tf_xlnet import (
TF_XLNET_PRETRAINED_MODEL_ARCHIVE_LIST,
TFXLNetForMultipleChoice,
TFXLNetForQuestionAnsweringSimple,
TFXLNetForSequenceClassification,
TFXLNetForTokenClassification,
TFXLNetLMHeadModel,
TFXLNetMainLayer,
TFXLNetModel,
TFXLNetPreTrainedModel,
)
else:
import sys
snake_case__ : Optional[int] = _LazyModule(__name__, globals()["""__file__"""], _import_structure, module_spec=__spec__)
| 23 |
from __future__ import annotations
from sys import maxsize
from typing import Generic, TypeVar
snake_case__ : List[str] = TypeVar("""T""")
def _snake_case (__lowercase):
return (position - 1) // 2
def _snake_case (__lowercase):
return (2 * position) + 1
def _snake_case (__lowercase):
return (2 * position) + 2
class _a ( Generic[T] ):
"""simple docstring"""
def __init__( self ) -> None:
UpperCamelCase_ = []
UpperCamelCase_ = {}
UpperCamelCase_ = 0
def __len__( self ) -> int:
return self.elements
def __repr__( self ) -> str:
return str(self.heap )
def _UpperCAmelCase ( self ) -> bool:
# Check if the priority queue is empty
return self.elements == 0
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> None:
# Add an element with given priority to the queue
self.heap.append((elem, weight) )
UpperCamelCase_ = self.elements
self.elements += 1
self._bubble_up(_UpperCAmelCase )
def _UpperCAmelCase ( self ) -> T:
# Remove and return the element with lowest weight (highest priority)
if self.elements > 1:
self._swap_nodes(0 , self.elements - 1 )
UpperCamelCase_ , UpperCamelCase_ = self.heap.pop()
del self.position_map[elem]
self.elements -= 1
if self.elements > 0:
UpperCamelCase_ , UpperCamelCase_ = self.heap[0]
self._bubble_down(_UpperCAmelCase )
return elem
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> None:
# Update the weight of the given key
UpperCamelCase_ = self.position_map[elem]
UpperCamelCase_ = (elem, weight)
if position > 0:
UpperCamelCase_ = get_parent_position(_UpperCAmelCase )
UpperCamelCase_ , UpperCamelCase_ = self.heap[parent_position]
if parent_weight > weight:
self._bubble_up(_UpperCAmelCase )
else:
self._bubble_down(_UpperCAmelCase )
else:
self._bubble_down(_UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase ) -> None:
# Place a node at the proper position (upward movement) [to be used internally
# only]
UpperCamelCase_ = self.position_map[elem]
if curr_pos == 0:
return None
UpperCamelCase_ = get_parent_position(_UpperCAmelCase )
UpperCamelCase_ , UpperCamelCase_ = self.heap[curr_pos]
UpperCamelCase_ , UpperCamelCase_ = self.heap[parent_position]
if parent_weight > weight:
self._swap_nodes(_UpperCAmelCase , _UpperCAmelCase )
return self._bubble_up(_UpperCAmelCase )
return None
def _UpperCAmelCase ( self , _UpperCAmelCase ) -> None:
# Place a node at the proper position (downward movement) [to be used
# internally only]
UpperCamelCase_ = self.position_map[elem]
UpperCamelCase_ , UpperCamelCase_ = self.heap[curr_pos]
UpperCamelCase_ = get_child_left_position(_UpperCAmelCase )
UpperCamelCase_ = get_child_right_position(_UpperCAmelCase )
if child_left_position < self.elements and child_right_position < self.elements:
UpperCamelCase_ , UpperCamelCase_ = self.heap[child_left_position]
UpperCamelCase_ , UpperCamelCase_ = self.heap[child_right_position]
if child_right_weight < child_left_weight and child_right_weight < weight:
self._swap_nodes(_UpperCAmelCase , _UpperCAmelCase )
return self._bubble_down(_UpperCAmelCase )
if child_left_position < self.elements:
UpperCamelCase_ , UpperCamelCase_ = self.heap[child_left_position]
if child_left_weight < weight:
self._swap_nodes(_UpperCAmelCase , _UpperCAmelCase )
return self._bubble_down(_UpperCAmelCase )
else:
return None
if child_right_position < self.elements:
UpperCamelCase_ , UpperCamelCase_ = self.heap[child_right_position]
if child_right_weight < weight:
self._swap_nodes(_UpperCAmelCase , _UpperCAmelCase )
return self._bubble_down(_UpperCAmelCase )
return None
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> None:
# Swap the nodes at the given positions
UpperCamelCase_ = self.heap[nodea_pos][0]
UpperCamelCase_ = self.heap[nodea_pos][0]
UpperCamelCase_ , UpperCamelCase_ = (
self.heap[nodea_pos],
self.heap[nodea_pos],
)
UpperCamelCase_ = nodea_pos
UpperCamelCase_ = nodea_pos
class _a ( Generic[T] ):
"""simple docstring"""
def __init__( self ) -> None:
UpperCamelCase_ = {}
UpperCamelCase_ = 0
def __repr__( self ) -> str:
return str(self.connections )
def __len__( self ) -> int:
return self.nodes
def _UpperCAmelCase ( self , _UpperCAmelCase ) -> None:
# Add a node in the graph if it is not in the graph
if node not in self.connections:
UpperCamelCase_ = {}
self.nodes += 1
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase ) -> None:
# Add an edge between 2 nodes in the graph
self.add_node(_UpperCAmelCase )
self.add_node(_UpperCAmelCase )
UpperCamelCase_ = weight
UpperCamelCase_ = weight
def _snake_case (__lowercase , ):
UpperCamelCase_ = {node: maxsize for node in graph.connections}
UpperCamelCase_ = {node: None for node in graph.connections}
UpperCamelCase_ = MinPriorityQueue()
for node, weight in dist.items():
priority_queue.push(__lowercase , __lowercase)
if priority_queue.is_empty():
return dist, parent
# initialization
UpperCamelCase_ = priority_queue.extract_min()
UpperCamelCase_ = 0
for neighbour in graph.connections[node]:
if dist[neighbour] > dist[node] + graph.connections[node][neighbour]:
UpperCamelCase_ = dist[node] + graph.connections[node][neighbour]
priority_queue.update_key(__lowercase , dist[neighbour])
UpperCamelCase_ = node
# running prim's algorithm
while not priority_queue.is_empty():
UpperCamelCase_ = priority_queue.extract_min()
for neighbour in graph.connections[node]:
if dist[neighbour] > dist[node] + graph.connections[node][neighbour]:
UpperCamelCase_ = dist[node] + graph.connections[node][neighbour]
priority_queue.update_key(__lowercase , dist[neighbour])
UpperCamelCase_ = node
return dist, parent
| 23 | 1 |
import doctest
from collections import deque
import numpy as np
class _a :
"""simple docstring"""
def __init__( self ) -> None:
UpperCamelCase_ = [2, 1, 2, -1]
UpperCamelCase_ = [1, 2, 3, 4]
def _UpperCAmelCase ( self ) -> list[float]:
UpperCamelCase_ = len(self.first_signal )
UpperCamelCase_ = len(self.second_signal )
UpperCamelCase_ = max(_UpperCAmelCase , _UpperCAmelCase )
# create a zero matrix of max_length x max_length
UpperCamelCase_ = [[0] * max_length for i in range(_UpperCAmelCase )]
# fills the smaller signal with zeros to make both signals of same length
if length_first_signal < length_second_signal:
self.first_signal += [0] * (max_length - length_first_signal)
elif length_first_signal > length_second_signal:
self.second_signal += [0] * (max_length - length_second_signal)
for i in range(_UpperCAmelCase ):
UpperCamelCase_ = deque(self.second_signal )
rotated_signal.rotate(_UpperCAmelCase )
for j, item in enumerate(_UpperCAmelCase ):
matrix[i][j] += item
# multiply the matrix with the first signal
UpperCamelCase_ = np.matmul(np.transpose(_UpperCAmelCase ) , np.transpose(self.first_signal ) )
# rounding-off to two decimal places
return [round(_UpperCAmelCase , 2 ) for i in final_signal]
if __name__ == "__main__":
doctest.testmod()
| 23 |
from __future__ import annotations
import sys
from collections import deque
from typing import Generic, TypeVar
snake_case__ : Dict = TypeVar("""T""")
class _a ( Generic[T] ):
"""simple docstring"""
A_ = 42 # Cache store of keys
A_ = 42 # References of the keys in cache
A_ = 10 # Maximum capacity of cache
def __init__( self , _UpperCAmelCase ) -> None:
UpperCamelCase_ = deque()
UpperCamelCase_ = set()
if not n:
UpperCamelCase_ = sys.maxsize
elif n < 0:
raise ValueError('n should be an integer greater than 0.' )
else:
UpperCamelCase_ = n
def _UpperCAmelCase ( self , _UpperCAmelCase ) -> None:
if x not in self.key_reference:
if len(self.dq_store ) == LRUCache._MAX_CAPACITY:
UpperCamelCase_ = self.dq_store.pop()
self.key_reference.remove(_UpperCAmelCase )
else:
self.dq_store.remove(_UpperCAmelCase )
self.dq_store.appendleft(_UpperCAmelCase )
self.key_reference.add(_UpperCAmelCase )
def _UpperCAmelCase ( self ) -> None:
for k in self.dq_store:
print(_UpperCAmelCase )
def __repr__( self ) -> str:
return f"""LRUCache({self._MAX_CAPACITY}) => {list(self.dq_store )}"""
if __name__ == "__main__":
import doctest
doctest.testmod()
snake_case__ : LRUCache[str | int] = LRUCache(4)
lru_cache.refer("""A""")
lru_cache.refer(2)
lru_cache.refer(3)
lru_cache.refer("""A""")
lru_cache.refer(4)
lru_cache.refer(5)
lru_cache.display()
print(lru_cache)
assert str(lru_cache) == "LRUCache(4) => [5, 4, 'A', 3]"
| 23 | 1 |
import sys
import turtle
def _snake_case (__lowercase , __lowercase):
return (pa[0] + pa[0]) / 2, (pa[1] + pa[1]) / 2
def _snake_case (__lowercase , __lowercase , __lowercase , __lowercase , ):
my_pen.up()
my_pen.goto(vertexa[0] , vertexa[1])
my_pen.down()
my_pen.goto(vertexa[0] , vertexa[1])
my_pen.goto(vertexa[0] , vertexa[1])
my_pen.goto(vertexa[0] , vertexa[1])
if depth == 0:
return
triangle(__lowercase , get_mid(__lowercase , __lowercase) , get_mid(__lowercase , __lowercase) , depth - 1)
triangle(__lowercase , get_mid(__lowercase , __lowercase) , get_mid(__lowercase , __lowercase) , depth - 1)
triangle(__lowercase , get_mid(__lowercase , __lowercase) , get_mid(__lowercase , __lowercase) , depth - 1)
if __name__ == "__main__":
if len(sys.argv) != 2:
raise ValueError(
"""Correct format for using this script: """
"""python fractals.py <int:depth_for_fractal>"""
)
snake_case__ : Tuple = turtle.Turtle()
my_pen.ht()
my_pen.speed(5)
my_pen.pencolor("""red""")
snake_case__ : Optional[Any] = [(-1_7_5, -1_2_5), (0, 1_7_5), (1_7_5, -1_2_5)] # vertices of triangle
triangle(vertices[0], vertices[1], vertices[2], int(sys.argv[1]))
| 23 |
import numpy as np
def _snake_case (__lowercase):
return 1 / (1 + np.exp(-vector))
def _snake_case (__lowercase):
return vector * sigmoid(__lowercase)
if __name__ == "__main__":
import doctest
doctest.testmod()
| 23 | 1 |
import argparse
from pathlib import Path
from transformers import AutoConfig, AutoTokenizer, RagConfig, RagSequenceForGeneration, RagTokenForGeneration
def _snake_case (__lowercase , __lowercase , __lowercase , __lowercase , __lowercase = None , __lowercase = None , __lowercase = None , ):
if config_name_or_path is None:
UpperCamelCase_ = 'facebook/rag-token-base' if model_type == 'rag_token' else 'facebook/rag-sequence-base'
if generator_tokenizer_name_or_path is None:
UpperCamelCase_ = generator_name_or_path
if question_encoder_tokenizer_name_or_path is None:
UpperCamelCase_ = question_encoder_name_or_path
UpperCamelCase_ = RagTokenForGeneration if model_type == 'rag_token' else RagSequenceForGeneration
# Save model.
UpperCamelCase_ = RagConfig.from_pretrained(__lowercase)
UpperCamelCase_ = AutoConfig.from_pretrained(__lowercase)
UpperCamelCase_ = AutoConfig.from_pretrained(__lowercase)
UpperCamelCase_ = gen_config
UpperCamelCase_ = question_encoder_config
UpperCamelCase_ = model_class.from_pretrained_question_encoder_generator(
__lowercase , __lowercase , config=__lowercase)
rag_model.save_pretrained(__lowercase)
# Sanity check.
model_class.from_pretrained(__lowercase)
# Save tokenizers.
UpperCamelCase_ = AutoTokenizer.from_pretrained(__lowercase)
gen_tokenizer.save_pretrained(dest_dir / 'generator_tokenizer/')
UpperCamelCase_ = AutoTokenizer.from_pretrained(__lowercase)
question_encoder_tokenizer.save_pretrained(dest_dir / 'question_encoder_tokenizer/')
if __name__ == "__main__":
snake_case__ : Union[str, Any] = argparse.ArgumentParser()
parser.add_argument(
"""--model_type""",
choices=["""rag_sequence""", """rag_token"""],
required=True,
type=str,
help="""RAG model type: rag_sequence, rag_token""",
)
parser.add_argument("""--dest""", type=str, required=True, help="""Path to the output checkpoint directory.""")
parser.add_argument("""--generator_name_or_path""", type=str, required=True, help="""Generator model identifier""")
parser.add_argument(
"""--question_encoder_name_or_path""", type=str, required=True, help="""Question encoder model identifier"""
)
parser.add_argument(
"""--generator_tokenizer_name_or_path""",
type=str,
help="""Generator tokenizer identifier, if not specified, resolves to ``generator_name_or_path``""",
)
parser.add_argument(
"""--question_encoder_tokenizer_name_or_path""",
type=str,
help="""Question encoder tokenizer identifier, if not specified, resolves to ``question_encoder_name_or_path``""",
)
parser.add_argument(
"""--config_name_or_path""",
type=str,
help=(
"""Identifier of the model config to use, if not provided, resolves to a base config for a given"""
""" ``model_type``"""
),
)
snake_case__ : int = parser.parse_args()
snake_case__ : str = Path(args.dest)
dest_dir.mkdir(exist_ok=True)
consolidate(
args.model_type,
args.generator_name_or_path,
args.question_encoder_name_or_path,
dest_dir,
args.config_name_or_path,
args.generator_tokenizer_name_or_path,
args.question_encoder_tokenizer_name_or_path,
)
| 23 |
import math
from datetime import datetime, timedelta
def _snake_case (__lowercase):
UpperCamelCase_ = year % 19
UpperCamelCase_ = year % 4
UpperCamelCase_ = year % 7
UpperCamelCase_ = math.floor(year / 100)
UpperCamelCase_ = math.floor((13 + 8 * leap_day_inhibits) / 25)
UpperCamelCase_ = leap_day_inhibits / 4
UpperCamelCase_ = (
15 - lunar_orbit_correction + leap_day_inhibits - leap_day_reinstall_number
) % 30
UpperCamelCase_ = (4 + leap_day_inhibits - leap_day_reinstall_number) % 7
# days to be added to March 21
UpperCamelCase_ = (19 * metonic_cycle + secular_moon_shift) % 30
# PHM -> Paschal Full Moon
UpperCamelCase_ = (
2 * julian_leap_year
+ 4 * non_leap_year
+ 6 * days_to_add
+ century_starting_point
) % 7
if days_to_add == 29 and days_from_phm_to_sunday == 6:
return datetime(__lowercase , 4 , 19)
elif days_to_add == 28 and days_from_phm_to_sunday == 6:
return datetime(__lowercase , 4 , 18)
else:
return datetime(__lowercase , 3 , 22) + timedelta(
days=int(days_to_add + days_from_phm_to_sunday))
if __name__ == "__main__":
for year in (1_9_9_4, 2_0_0_0, 2_0_1_0, 2_0_2_1, 2_0_2_3):
snake_case__ : Dict = """will be""" if year > datetime.now().year else """was"""
print(f'Easter in {year} {tense} {gauss_easter(year)}')
| 23 | 1 |
import os
import tempfile
import unittest
from transformers.models.marian.convert_marian_tatoeba_to_pytorch import DEFAULT_REPO, TatoebaConverter
from transformers.testing_utils import slow
from transformers.utils import cached_property
@unittest.skipUnless(os.path.exists(UpperCAmelCase__ ) , """Tatoeba directory does not exist.""" )
class _a ( unittest.TestCase ):
"""simple docstring"""
@cached_property
def _UpperCAmelCase ( self ) -> str:
UpperCamelCase_ = tempfile.mkdtemp()
return TatoebaConverter(save_dir=_UpperCAmelCase )
@slow
def _UpperCAmelCase ( self ) -> Union[str, Any]:
self.resolver.convert_models(['heb-eng'] )
@slow
def _UpperCAmelCase ( self ) -> List[str]:
UpperCamelCase_ , UpperCamelCase_ = self.resolver.write_model_card('opus-mt-he-en' , dry_run=_UpperCAmelCase )
assert mmeta["long_pair"] == "heb-eng"
| 23 |
import requests
def _snake_case (__lowercase , __lowercase):
UpperCamelCase_ = {'Content-Type': 'application/json'}
UpperCamelCase_ = requests.post(__lowercase , json={'text': message_body} , headers=__lowercase)
if response.status_code != 200:
UpperCamelCase_ = (
'Request to slack returned an error '
f"""{response.status_code}, the response is:\n{response.text}"""
)
raise ValueError(__lowercase)
if __name__ == "__main__":
# Set the slack url to the one provided by Slack when you create the webhook at
# https://my.slack.com/services/new/incoming-webhook/
send_slack_message("""<YOUR MESSAGE BODY>""", """<SLACK CHANNEL URL>""")
| 23 | 1 |
import json
from typing import List, Optional, Tuple
from tokenizers import normalizers
from ....tokenization_utils_fast import PreTrainedTokenizerFast
from ....utils import logging
from .tokenization_retribert import RetriBertTokenizer
snake_case__ : Optional[Any] = logging.get_logger(__name__)
snake_case__ : Dict = {"""vocab_file""": """vocab.txt""", """tokenizer_file""": """tokenizer.json"""}
snake_case__ : Union[str, Any] = {
"""vocab_file""": {
"""yjernite/retribert-base-uncased""": (
"""https://huggingface.co/yjernite/retribert-base-uncased/resolve/main/vocab.txt"""
),
},
"""tokenizer_file""": {
"""yjernite/retribert-base-uncased""": (
"""https://huggingface.co/yjernite/retribert-base-uncased/resolve/main/tokenizer.json"""
),
},
}
snake_case__ : List[Any] = {
"""yjernite/retribert-base-uncased""": 5_1_2,
}
snake_case__ : List[str] = {
"""yjernite/retribert-base-uncased""": {"""do_lower_case""": True},
}
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
A_ = VOCAB_FILES_NAMES
A_ = PRETRAINED_VOCAB_FILES_MAP
A_ = PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES
A_ = PRETRAINED_INIT_CONFIGURATION
A_ = RetriBertTokenizer
A_ = ["""input_ids""", """attention_mask"""]
def __init__( self , _UpperCAmelCase=None , _UpperCAmelCase=None , _UpperCAmelCase=True , _UpperCAmelCase="[UNK]" , _UpperCAmelCase="[SEP]" , _UpperCAmelCase="[PAD]" , _UpperCAmelCase="[CLS]" , _UpperCAmelCase="[MASK]" , _UpperCAmelCase=True , _UpperCAmelCase=None , **_UpperCAmelCase , ) -> int:
super().__init__(
_UpperCAmelCase , tokenizer_file=_UpperCAmelCase , do_lower_case=_UpperCAmelCase , unk_token=_UpperCAmelCase , sep_token=_UpperCAmelCase , pad_token=_UpperCAmelCase , cls_token=_UpperCAmelCase , mask_token=_UpperCAmelCase , tokenize_chinese_chars=_UpperCAmelCase , strip_accents=_UpperCAmelCase , **_UpperCAmelCase , )
UpperCamelCase_ = json.loads(self.backend_tokenizer.normalizer.__getstate__() )
if (
normalizer_state.get('lowercase' , _UpperCAmelCase ) != do_lower_case
or normalizer_state.get('strip_accents' , _UpperCAmelCase ) != strip_accents
or normalizer_state.get('handle_chinese_chars' , _UpperCAmelCase ) != tokenize_chinese_chars
):
UpperCamelCase_ = getattr(_UpperCAmelCase , normalizer_state.pop('type' ) )
UpperCamelCase_ = do_lower_case
UpperCamelCase_ = strip_accents
UpperCamelCase_ = tokenize_chinese_chars
UpperCamelCase_ = normalizer_class(**_UpperCAmelCase )
UpperCamelCase_ = do_lower_case
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase=None ) -> Union[str, Any]:
UpperCamelCase_ = [self.cls_token_id] + token_ids_a + [self.sep_token_id]
if token_ids_a:
output += token_ids_a + [self.sep_token_id]
return output
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase = None ) -> List[int]:
UpperCamelCase_ = [self.sep_token_id]
UpperCamelCase_ = [self.cls_token_id]
if token_ids_a is None:
return len(cls + token_ids_a + sep ) * [0]
return len(cls + token_ids_a + sep ) * [0] + len(token_ids_a + sep ) * [1]
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase = None ) -> Tuple[str]:
UpperCamelCase_ = self._tokenizer.model.save(_UpperCAmelCase , name=_UpperCAmelCase )
return tuple(_UpperCAmelCase )
| 23 |
import re
from pathlib import Path
from unittest import TestCase
import pytest
@pytest.mark.integration
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
def _UpperCAmelCase ( self , _UpperCAmelCase ) -> Dict:
with open(_UpperCAmelCase , encoding='utf-8' ) as input_file:
UpperCamelCase_ = re.compile(R'(?!.*\b(?:encoding|rb|w|wb|w+|wb+|ab|ab+)\b)(?<=\s)(open)\((.*)\)' )
UpperCamelCase_ = input_file.read()
UpperCamelCase_ = regexp.search(_UpperCAmelCase )
return match
def _UpperCAmelCase ( self , _UpperCAmelCase ) -> Dict:
with open(_UpperCAmelCase , encoding='utf-8' ) as input_file:
UpperCamelCase_ = re.compile(R'#[^\r\n]*print\(|\"[^\r\n]*print\(|\"\"\".*?print\(.*?\"\"\"|(print\()' , re.DOTALL )
UpperCamelCase_ = input_file.read()
# use `re.finditer` to handle the case where the ignored groups would be matched first by `re.search`
UpperCamelCase_ = regexp.finditer(_UpperCAmelCase )
UpperCamelCase_ = [match for match in matches if match is not None and match.group(1 ) is not None]
return matches[0] if matches else None
def _UpperCAmelCase ( self ) -> List[str]:
UpperCamelCase_ = Path('./datasets' )
UpperCamelCase_ = list(dataset_paths.absolute().glob('**/*.py' ) )
for dataset in dataset_files:
if self._no_encoding_on_file_open(str(_UpperCAmelCase ) ):
raise AssertionError(f"""open(...) must use utf-8 encoding in {dataset}""" )
def _UpperCAmelCase ( self ) -> Optional[int]:
UpperCamelCase_ = Path('./datasets' )
UpperCamelCase_ = list(dataset_paths.absolute().glob('**/*.py' ) )
for dataset in dataset_files:
if self._no_print_statements(str(_UpperCAmelCase ) ):
raise AssertionError(f"""print statement found in {dataset}. Use datasets.logger/logging instead.""" )
| 23 | 1 |
import warnings
from ...utils import logging
from .image_processing_layoutlmva import LayoutLMvaImageProcessor
snake_case__ : Optional[Any] = logging.get_logger(__name__)
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
def __init__( self , *_UpperCAmelCase , **_UpperCAmelCase ) -> None:
warnings.warn(
'The class LayoutLMv2FeatureExtractor is deprecated and will be removed in version 5 of Transformers.'
' Please use LayoutLMv2ImageProcessor instead.' , _UpperCAmelCase , )
super().__init__(*_UpperCAmelCase , **_UpperCAmelCase )
| 23 |
import argparse
import random
import joblib
import numpy as np
import torch
from igf.igf import (
SecondaryLearner,
collect_objective_set,
compute_perplexity,
generate_datasets,
load_gpta,
recopy_gpta,
set_seed,
train_secondary_learner,
)
from torch.utils.data import DataLoader, RandomSampler
from transformers import GPTaLMHeadModel
def _snake_case (__lowercase=32 , __lowercase=10 , __lowercase=100 , __lowercase=1026 , __lowercase=True , __lowercase="data/tokenized_stories_train_wikitext103.jbl" , __lowercase="igf_context_pairs.jbl" , ):
set_seed(3)
# generate train_data and objective_set
UpperCamelCase_ , UpperCamelCase_ = generate_datasets(
__lowercase , __lowercase , number=__lowercase , min_len=1026 , trim=__lowercase)
# keeps model same across runs
set_seed(4)
# model, lm_optimizer, lm_scheduler = recopy_gpt2(model, device, max_steps) # store original model weights
# can we train on GPU?
UpperCamelCase_ = torch.device('cuda:0' if torch.cuda.is_available() else 'cpu')
# load pretrained model
UpperCamelCase_ = load_gpta('gpt2').to(__lowercase)
print('computing perplexity on objective set')
UpperCamelCase_ = compute_perplexity(__lowercase , __lowercase , __lowercase).item()
print('perplexity on objective set:' , __lowercase)
# collect igf pairs and save to file demo.jbl
collect_objective_set(__lowercase , __lowercase , __lowercase , __lowercase , __lowercase , __lowercase , __lowercase , __lowercase)
# clean up, delete model and data we don't need anymore
del model, train_data, objective_set
torch.cuda.empty_cache()
def _snake_case (__lowercase , __lowercase=15 , __lowercase=128 , __lowercase=100 , __lowercase="igf_model.pt" , ):
set_seed(42)
# Load pre-trained model
UpperCamelCase_ = GPTaLMHeadModel.from_pretrained('gpt2')
# Initialize secondary learner to use embedding weights of model
UpperCamelCase_ = SecondaryLearner(__lowercase)
# Train secondary learner
UpperCamelCase_ = train_secondary_learner(
__lowercase , __lowercase , max_epochs=__lowercase , batch_size=__lowercase , eval_freq=100 , igf_model_path=__lowercase , )
del model, secondary_learner_train_data
torch.cuda.empty_cache()
return secondary_learner
def _snake_case (__lowercase , __lowercase , __lowercase , __lowercase=32 , __lowercase=1000 , __lowercase=16 , __lowercase=1.0 , __lowercase=recopy_gpta , __lowercase=None , __lowercase=10 , __lowercase="gpt2_finetuned.pt" , ):
UpperCamelCase_ = torch.device('cuda:0' if torch.cuda.is_available() else 'cpu')
UpperCamelCase_ = RandomSampler(__lowercase)
UpperCamelCase_ = DataLoader(__lowercase , sampler=__lowercase)
UpperCamelCase_ = max_steps // (len(__lowercase)) + 1
UpperCamelCase_ = 0
UpperCamelCase_ = torch.zeros((1, context_len) , dtype=torch.long , device=__lowercase)
UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ = recopy_model(__lowercase , __lowercase , __lowercase)
model.train()
if secondary_learner is not None:
secondary_learner.to(__lowercase)
secondary_learner.eval()
UpperCamelCase_ = []
UpperCamelCase_ = 0
UpperCamelCase_ = []
UpperCamelCase_ = []
# Compute the performance of the transformer model at the beginning
UpperCamelCase_ = compute_perplexity(__lowercase , __lowercase , __lowercase)
test_perps.append(__lowercase)
print('Test perplexity, step' , __lowercase , ':' , __lowercase)
for epoch in range(int(__lowercase)):
for step, example in enumerate(__lowercase):
torch.cuda.empty_cache()
UpperCamelCase_ = random.randint(0 , example.size(2) - context_len - 1)
UpperCamelCase_ = example[0, 0, start : start + context_len]
lm_optimizer.zero_grad()
UpperCamelCase_ = model(__lowercase , labels=__lowercase)
UpperCamelCase_ = True
if secondary_learner is not None:
UpperCamelCase_ = secondary_learner.forward(
torch.tensor(__lowercase , dtype=torch.long , device=__lowercase).unsqueeze(0))[0].item()
observed_qs.append(float(__lowercase))
# Here we implement the simple non-constant threshold for the predicted IG(X) value
# We will decay the selectivity of our secondary learner filter from
# 1 standard deviation above average to 1 below average after 10 batches.
if global_step == 10:
UpperCamelCase_ = -1
if predicted_q < threshold:
UpperCamelCase_ = False
# If we passed the filter, add the context to the batch!
if do_backprop:
contexts.append(np.array(context.cpu()))
UpperCamelCase_ = outputs[0]
lm_loss.backward()
examples += 1
del outputs
# Once the batch is filled with enough contexts, backprop on the batch.
if examples == batch_size:
torch.cuda.empty_cache()
UpperCamelCase_ = 0
# Do LM backprop
torch.nn.utils.clip_grad_norm_(model.parameters() , 3.0)
lm_optimizer.step()
lm_scheduler.step() # Update learning rate schedule
global_step += 1
# Compute the performance of the transformer model at this batch
if global_step % eval_interval == 0:
UpperCamelCase_ = compute_perplexity(__lowercase , __lowercase , __lowercase)
test_perps.append(__lowercase)
print('Test perplexity, step' , __lowercase , ':' , __lowercase)
# Break out of the loop after 60 batches
if max_steps > 0 and global_step > 60:
break
if max_steps > 0 and global_step > 60:
break
# save finetuned transformer model
torch.save(model.state_dict() , __lowercase)
torch.cuda.empty_cache()
# Do some cleaning up so we can reinitialize for the next run of this function
del lm_optimizer
del lm_scheduler
return model
def _snake_case ():
UpperCamelCase_ = argparse.ArgumentParser(description='Fine-tune a transformer model with IGF on a language modeling task')
# Required parameters
parser.add_argument(
'--data_dir' , default=__lowercase , type=__lowercase , required=__lowercase , help='The input data dir. Should contain data files for WikiText.' , )
parser.add_argument(
'--model_name_or_path' , default=__lowercase , type=__lowercase , required=__lowercase , help='Path to pretrained model or model identifier from huggingface.co/models' , )
parser.add_argument(
'--data_file' , type=__lowercase , default=__lowercase , help=(
'A jbl file containing tokenized data which can be split as objective dataset, '
'train_dataset and test_dataset.'
) , )
parser.add_argument(
'--igf_data_file' , type=__lowercase , default=__lowercase , help='A jbl file containing the context and information gain pairs to train secondary learner.' , )
parser.add_argument(
'--output_dir' , default=__lowercase , type=__lowercase , required=__lowercase , help='The output directory where the final fine-tuned model is stored.' , )
parser.add_argument(
'--tokenizer_name' , default=__lowercase , type=__lowercase , help='Pretrained tokenizer name or path if not the same as model_name' , )
parser.add_argument('--seed' , type=__lowercase , default=__lowercase , help='A seed for reproducible training.')
parser.add_argument(
'--context_len' , default=32 , type=__lowercase , help=(
'The maximum total input sequence length after tokenization. Sequences longer '
'than this will be truncated, sequences shorter will be padded.'
) , )
parser.add_argument(
'--size_objective_set' , default=100 , type=__lowercase , help='number of articles that are long enough to be used as our objective set' , )
parser.add_argument(
'--eval_freq' , default=100 , type=__lowercase , help='secondary model evaluation is triggered at eval_freq')
parser.add_argument('--max_steps' , default=1000 , type=__lowercase , help='To calculate training epochs')
parser.add_argument(
'--secondary_learner_batch_size' , default=128 , type=__lowercase , help='batch size of training data for secondary learner' , )
parser.add_argument(
'--batch_size' , default=16 , type=__lowercase , help='batch size of training data of language model(gpt2) ')
parser.add_argument(
'--eval_interval' , default=10 , type=__lowercase , help=(
'decay the selectivity of our secondary learner filter from'
'1 standard deviation above average to 1 below average after 10 batches'
) , )
parser.add_argument(
'--number' , default=100 , type=__lowercase , help='The number of examples split to be used as objective_set/test_data')
parser.add_argument(
'--min_len' , default=1026 , type=__lowercase , help='The minimum length of the article to be used as objective set')
parser.add_argument(
'--secondary_learner_max_epochs' , default=15 , type=__lowercase , help='number of epochs to train secondary learner')
parser.add_argument('--trim' , default=__lowercase , type=__lowercase , help='truncate the example if it exceeds context length')
parser.add_argument(
'--threshold' , default=1.0 , type=__lowercase , help=(
'The threshold value used by secondary learner to filter the train_data and allow only'
' informative data as input to the model'
) , )
parser.add_argument('--finetuned_model_name' , default='gpt2_finetuned.pt' , type=__lowercase , help='finetuned_model_name')
parser.add_argument(
'--recopy_model' , default=__lowercase , type=__lowercase , help='Reset the model to the original pretrained GPT-2 weights after each iteration' , )
# function calls
# Collecting *n* pairs of context and information gain(X, IG(X)) for training the secondary learner
generate_n_pairs(
context_len=32 , max_steps=10 , size_objective_set=100 , min_len=1026 , trim=__lowercase , data_file='data/tokenized_stories_train_wikitext103.jbl' , igf_data_file='igf_context_pairs.jbl' , )
# Load train data for secondary learner
UpperCamelCase_ = joblib.load('data/IGF_values.jbl')
# Train secondary learner
UpperCamelCase_ = training_secondary_learner(
__lowercase , secondary_learner_max_epochs=15 , secondary_learner_batch_size=128 , eval_freq=100 , igf_model_path='igf_model.pt' , )
# load pretrained gpt2 model
UpperCamelCase_ = GPTaLMHeadModel.from_pretrained('gpt2')
set_seed(42)
# Generate train and test data to train and evaluate gpt2 model
UpperCamelCase_ , UpperCamelCase_ = generate_datasets(
context_len=32 , file='data/tokenized_stories_train_wikitext103.jbl' , number=100 , min_len=1026 , trim=__lowercase)
# fine-tuning of the gpt2 model using igf (Information Gain Filtration)
finetune(
__lowercase , __lowercase , __lowercase , context_len=32 , max_steps=1000 , batch_size=16 , threshold=1.0 , recopy_model=__lowercase , secondary_learner=__lowercase , eval_interval=10 , finetuned_model_name='gpt2_finetuned.pt' , )
if __name__ == "__main__":
main()
| 23 | 1 |
import torch
import torch.nn as nn
from transformers import CLIPConfig, CLIPVisionModel, PreTrainedModel
from ...utils import logging
snake_case__ : Tuple = logging.get_logger(__name__)
def _snake_case (__lowercase , __lowercase):
UpperCamelCase_ = nn.functional.normalize(__lowercase)
UpperCamelCase_ = nn.functional.normalize(__lowercase)
return torch.mm(__lowercase , normalized_text_embeds.t())
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
A_ = CLIPConfig
A_ = ["""CLIPEncoderLayer"""]
def __init__( self , _UpperCAmelCase ) -> Dict:
super().__init__(_UpperCAmelCase )
UpperCamelCase_ = CLIPVisionModel(config.vision_config )
UpperCamelCase_ = nn.Linear(config.vision_config.hidden_size , config.projection_dim , bias=_UpperCAmelCase )
UpperCamelCase_ = nn.Parameter(torch.ones(17 , config.projection_dim ) , requires_grad=_UpperCAmelCase )
UpperCamelCase_ = nn.Parameter(torch.ones(3 , config.projection_dim ) , requires_grad=_UpperCAmelCase )
UpperCamelCase_ = nn.Parameter(torch.ones(17 ) , requires_grad=_UpperCAmelCase )
UpperCamelCase_ = nn.Parameter(torch.ones(3 ) , requires_grad=_UpperCAmelCase )
@torch.no_grad()
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> Optional[Any]:
UpperCamelCase_ = self.vision_model(_UpperCAmelCase )[1] # pooled_output
UpperCamelCase_ = self.visual_projection(_UpperCAmelCase )
# we always cast to float32 as this does not cause significant overhead and is compatible with bfloat16
UpperCamelCase_ = cosine_distance(_UpperCAmelCase , self.special_care_embeds ).cpu().float().numpy()
UpperCamelCase_ = cosine_distance(_UpperCAmelCase , self.concept_embeds ).cpu().float().numpy()
UpperCamelCase_ = []
UpperCamelCase_ = image_embeds.shape[0]
for i in range(_UpperCAmelCase ):
UpperCamelCase_ = {'special_scores': {}, 'special_care': [], 'concept_scores': {}, 'bad_concepts': []}
# increase this value to create a stronger `nfsw` filter
# at the cost of increasing the possibility of filtering benign images
UpperCamelCase_ = 0.0
for concept_idx in range(len(special_cos_dist[0] ) ):
UpperCamelCase_ = special_cos_dist[i][concept_idx]
UpperCamelCase_ = self.special_care_embeds_weights[concept_idx].item()
UpperCamelCase_ = round(concept_cos - concept_threshold + adjustment , 3 )
if result_img["special_scores"][concept_idx] > 0:
result_img["special_care"].append({concept_idx, result_img['special_scores'][concept_idx]} )
UpperCamelCase_ = 0.0_1
for concept_idx in range(len(cos_dist[0] ) ):
UpperCamelCase_ = cos_dist[i][concept_idx]
UpperCamelCase_ = self.concept_embeds_weights[concept_idx].item()
UpperCamelCase_ = round(concept_cos - concept_threshold + adjustment , 3 )
if result_img["concept_scores"][concept_idx] > 0:
result_img["bad_concepts"].append(_UpperCAmelCase )
result.append(_UpperCAmelCase )
UpperCamelCase_ = [len(res['bad_concepts'] ) > 0 for res in result]
return images, has_nsfw_concepts
@torch.no_grad()
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> List[str]:
UpperCamelCase_ = self.vision_model(_UpperCAmelCase )[1] # pooled_output
UpperCamelCase_ = self.visual_projection(_UpperCAmelCase )
UpperCamelCase_ = cosine_distance(_UpperCAmelCase , self.special_care_embeds )
UpperCamelCase_ = cosine_distance(_UpperCAmelCase , self.concept_embeds )
# increase this value to create a stronger `nsfw` filter
# at the cost of increasing the possibility of filtering benign images
UpperCamelCase_ = 0.0
UpperCamelCase_ = special_cos_dist - self.special_care_embeds_weights + adjustment
# special_scores = special_scores.round(decimals=3)
UpperCamelCase_ = torch.any(special_scores > 0 , dim=1 )
UpperCamelCase_ = special_care * 0.0_1
UpperCamelCase_ = special_adjustment.unsqueeze(1 ).expand(-1 , cos_dist.shape[1] )
UpperCamelCase_ = (cos_dist - self.concept_embeds_weights) + special_adjustment
# concept_scores = concept_scores.round(decimals=3)
UpperCamelCase_ = torch.any(concept_scores > 0 , dim=1 )
return images, has_nsfw_concepts
| 23 |
from __future__ import annotations
import unittest
from transformers import AutoTokenizer, MBartConfig, is_tf_available
from transformers.testing_utils import require_sentencepiece, require_tf, require_tokenizers, slow
from transformers.utils import cached_property
from ...test_configuration_common import ConfigTester
from ...test_modeling_tf_common import TFModelTesterMixin, ids_tensor
from ...test_pipeline_mixin import PipelineTesterMixin
if is_tf_available():
import tensorflow as tf
from transformers import TFAutoModelForSeqaSeqLM, TFMBartForConditionalGeneration, TFMBartModel
@require_tf
class _a :
"""simple docstring"""
A_ = MBartConfig
A_ = {}
A_ = """gelu"""
def __init__( self , _UpperCAmelCase , _UpperCAmelCase=13 , _UpperCAmelCase=7 , _UpperCAmelCase=True , _UpperCAmelCase=False , _UpperCAmelCase=99 , _UpperCAmelCase=32 , _UpperCAmelCase=2 , _UpperCAmelCase=4 , _UpperCAmelCase=37 , _UpperCAmelCase=0.1 , _UpperCAmelCase=0.1 , _UpperCAmelCase=20 , _UpperCAmelCase=2 , _UpperCAmelCase=1 , _UpperCAmelCase=0 , ) -> Union[str, Any]:
UpperCamelCase_ = parent
UpperCamelCase_ = batch_size
UpperCamelCase_ = seq_length
UpperCamelCase_ = is_training
UpperCamelCase_ = use_labels
UpperCamelCase_ = vocab_size
UpperCamelCase_ = hidden_size
UpperCamelCase_ = num_hidden_layers
UpperCamelCase_ = num_attention_heads
UpperCamelCase_ = intermediate_size
UpperCamelCase_ = hidden_dropout_prob
UpperCamelCase_ = attention_probs_dropout_prob
UpperCamelCase_ = max_position_embeddings
UpperCamelCase_ = eos_token_id
UpperCamelCase_ = pad_token_id
UpperCamelCase_ = bos_token_id
def _UpperCAmelCase ( self ) -> List[Any]:
UpperCamelCase_ = ids_tensor([self.batch_size, self.seq_length - 1] , self.vocab_size )
UpperCamelCase_ = tf.expand_dims(tf.constant([self.eos_token_id] * self.batch_size ) , 1 )
UpperCamelCase_ = tf.concat([input_ids, eos_tensor] , axis=1 )
UpperCamelCase_ = ids_tensor([self.batch_size, self.seq_length] , self.vocab_size )
UpperCamelCase_ = self.config_cls(
vocab_size=self.vocab_size , d_model=self.hidden_size , encoder_layers=self.num_hidden_layers , decoder_layers=self.num_hidden_layers , encoder_attention_heads=self.num_attention_heads , decoder_attention_heads=self.num_attention_heads , encoder_ffn_dim=self.intermediate_size , decoder_ffn_dim=self.intermediate_size , dropout=self.hidden_dropout_prob , attention_dropout=self.attention_probs_dropout_prob , max_position_embeddings=self.max_position_embeddings , eos_token_ids=[2] , bos_token_id=self.bos_token_id , pad_token_id=self.pad_token_id , decoder_start_token_id=self.pad_token_id , **self.config_updates , )
UpperCamelCase_ = prepare_mbart_inputs_dict(_UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase )
return config, inputs_dict
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> List[str]:
UpperCamelCase_ = TFMBartModel(config=_UpperCAmelCase ).get_decoder()
UpperCamelCase_ = inputs_dict['input_ids']
UpperCamelCase_ = input_ids[:1, :]
UpperCamelCase_ = inputs_dict['attention_mask'][:1, :]
UpperCamelCase_ = inputs_dict['head_mask']
UpperCamelCase_ = 1
# first forward pass
UpperCamelCase_ = model(_UpperCAmelCase , attention_mask=_UpperCAmelCase , head_mask=_UpperCAmelCase , use_cache=_UpperCAmelCase )
UpperCamelCase_ , UpperCamelCase_ = outputs.to_tuple()
UpperCamelCase_ = past_key_values[1]
def _snake_case (__lowercase , __lowercase , __lowercase , __lowercase=None , __lowercase=None , __lowercase=None , __lowercase=None , __lowercase=None , ):
if attention_mask is None:
UpperCamelCase_ = tf.cast(tf.math.not_equal(__lowercase , config.pad_token_id) , tf.inta)
if decoder_attention_mask is None:
UpperCamelCase_ = tf.concat(
[
tf.ones(decoder_input_ids[:, :1].shape , dtype=tf.inta),
tf.cast(tf.math.not_equal(decoder_input_ids[:, 1:] , config.pad_token_id) , tf.inta),
] , axis=-1 , )
if head_mask is None:
UpperCamelCase_ = tf.ones((config.encoder_layers, config.encoder_attention_heads))
if decoder_head_mask is None:
UpperCamelCase_ = tf.ones((config.decoder_layers, config.decoder_attention_heads))
if cross_attn_head_mask is None:
UpperCamelCase_ = tf.ones((config.decoder_layers, config.decoder_attention_heads))
return {
"input_ids": input_ids,
"decoder_input_ids": decoder_input_ids,
"attention_mask": attention_mask,
"decoder_attention_mask": decoder_attention_mask,
"head_mask": head_mask,
"decoder_head_mask": decoder_head_mask,
"cross_attn_head_mask": cross_attn_head_mask,
}
@require_tf
class _a ( UpperCAmelCase__ , UpperCAmelCase__ , unittest.TestCase ):
"""simple docstring"""
A_ = (TFMBartForConditionalGeneration, TFMBartModel) if is_tf_available() else ()
A_ = (TFMBartForConditionalGeneration,) if is_tf_available() else ()
A_ = (
{
"""conversational""": TFMBartForConditionalGeneration,
"""feature-extraction""": TFMBartModel,
"""summarization""": TFMBartForConditionalGeneration,
"""text2text-generation""": TFMBartForConditionalGeneration,
"""translation""": TFMBartForConditionalGeneration,
}
if is_tf_available()
else {}
)
A_ = True
A_ = False
A_ = False
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase ) -> Tuple:
if pipeline_test_casse_name != "FeatureExtractionPipelineTests":
# Exception encountered when calling layer '...'
return True
return False
def _UpperCAmelCase ( self ) -> Optional[int]:
UpperCamelCase_ = TFMBartModelTester(self )
UpperCamelCase_ = ConfigTester(self , config_class=_UpperCAmelCase )
def _UpperCAmelCase ( self ) -> Optional[int]:
self.config_tester.run_common_tests()
def _UpperCAmelCase ( self ) -> List[Any]:
UpperCamelCase_ = self.model_tester.prepare_config_and_inputs_for_common()
self.model_tester.check_decoder_model_past_large_inputs(*_UpperCAmelCase )
@require_sentencepiece
@require_tokenizers
@require_tf
class _a ( unittest.TestCase ):
"""simple docstring"""
A_ = [
""" UN Chief Says There Is No Military Solution in Syria""",
]
A_ = [
"""Şeful ONU declară că nu există o soluţie militară în Siria""",
]
A_ = """facebook/mbart-large-en-ro"""
@cached_property
def _UpperCAmelCase ( self ) -> Any:
return AutoTokenizer.from_pretrained(self.model_name )
@cached_property
def _UpperCAmelCase ( self ) -> List[str]:
UpperCamelCase_ = TFAutoModelForSeqaSeqLM.from_pretrained(self.model_name )
return model
def _UpperCAmelCase ( self , **_UpperCAmelCase ) -> int:
UpperCamelCase_ = self.translate_src_text(**_UpperCAmelCase )
self.assertListEqual(self.expected_text , _UpperCAmelCase )
def _UpperCAmelCase ( self , **_UpperCAmelCase ) -> List[str]:
UpperCamelCase_ = self.tokenizer(self.src_text , **_UpperCAmelCase , return_tensors='tf' )
UpperCamelCase_ = self.model.generate(
model_inputs.input_ids , attention_mask=model_inputs.attention_mask , num_beams=2 )
UpperCamelCase_ = self.tokenizer.batch_decode(_UpperCAmelCase , skip_special_tokens=_UpperCAmelCase )
return generated_words
@slow
def _UpperCAmelCase ( self ) -> List[Any]:
self._assert_generated_batch_equal_expected()
| 23 | 1 |
from math import ceil
from typing import List, Optional, Union
import numpy as np
from ...audio_utils import mel_filter_bank, spectrogram, window_function
from ...feature_extraction_sequence_utils import BatchFeature, SequenceFeatureExtractor
from ...utils import TensorType, logging
snake_case__ : int = logging.get_logger(__name__)
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
A_ = ["""audio_values""", """audio_mask"""]
def __init__( self , _UpperCAmelCase=2048 , _UpperCAmelCase=1 , _UpperCAmelCase=[16, 16] , _UpperCAmelCase=128 , _UpperCAmelCase=44100 , _UpperCAmelCase=86 , _UpperCAmelCase=2048 , _UpperCAmelCase=0.0 , **_UpperCAmelCase , ) -> Optional[Any]:
super().__init__(
feature_size=_UpperCAmelCase , sampling_rate=_UpperCAmelCase , padding_value=_UpperCAmelCase , **_UpperCAmelCase , )
UpperCamelCase_ = spectrogram_length
UpperCamelCase_ = num_channels
UpperCamelCase_ = patch_size
UpperCamelCase_ = feature_size // self.patch_size[1]
UpperCamelCase_ = n_fft
UpperCamelCase_ = sampling_rate // hop_length_to_sampling_rate
UpperCamelCase_ = sampling_rate
UpperCamelCase_ = padding_value
UpperCamelCase_ = mel_filter_bank(
num_frequency_bins=1 + n_fft // 2 , num_mel_filters=_UpperCAmelCase , min_frequency=0.0 , max_frequency=2_2_0_5_0.0 , sampling_rate=_UpperCAmelCase , norm='slaney' , mel_scale='slaney' , ).T
def _UpperCAmelCase ( self , _UpperCAmelCase ) -> np.ndarray:
UpperCamelCase_ = spectrogram(
_UpperCAmelCase , window_function(self.n_fft , 'hann' ) , frame_length=self.n_fft , hop_length=self.hop_length , power=2.0 , mel_filters=self.mel_filters.T , log_mel='dB' , db_range=8_0.0 , )
UpperCamelCase_ = log_spec[:, :-1]
UpperCamelCase_ = log_spec - 2_0.0
UpperCamelCase_ = np.clip(log_spec / 4_0.0 , -2.0 , 0.0 ) + 1.0
return log_spec
def __call__( self , _UpperCAmelCase , _UpperCAmelCase = None , _UpperCAmelCase = True , _UpperCAmelCase = None , _UpperCAmelCase = False , _UpperCAmelCase = False , **_UpperCAmelCase , ) -> BatchFeature:
if sampling_rate is not None:
if sampling_rate != self.sampling_rate:
raise ValueError(
'This feature extractor is set to support sampling rate'
f""" of {self.sampling_rate}. Please make sure that the provided `raw_speech` input was sampled"""
f""" with {self.sampling_rate} and not {sampling_rate}.""" )
else:
logger.warning(
'It is strongly recommended to pass the `sampling_rate` argument to this function. '
'Failing to do so can result in silent errors that might be hard to debug.' )
UpperCamelCase_ = isinstance(_UpperCAmelCase , np.ndarray ) and len(raw_speech.shape ) > 1
if is_batched_numpy and len(raw_speech.shape ) > 2:
raise ValueError(f"""Only mono-channel audio is supported for input to {self}""" )
UpperCamelCase_ = is_batched_numpy or (
isinstance(_UpperCAmelCase , (list, tuple) ) and (isinstance(raw_speech[0] , (np.ndarray, tuple, list) ))
)
if is_batched:
UpperCamelCase_ = [np.asarray([speech] , dtype=np.floataa ).T for speech in raw_speech]
elif not is_batched and not isinstance(_UpperCAmelCase , np.ndarray ):
UpperCamelCase_ = np.asarray(_UpperCAmelCase , dtype=np.floataa )
elif isinstance(_UpperCAmelCase , np.ndarray ) and raw_speech.dtype is np.dtype(np.floataa ):
UpperCamelCase_ = raw_speech.astype(np.floataa )
# always return batch
if not is_batched:
UpperCamelCase_ = [np.asarray([raw_speech] ).T]
# Convert audio signals to log mel spectrograms, truncate by time axis
UpperCamelCase_ = [
self._np_extract_fbank_features(waveform.squeeze() ).T[: self.spectrogram_length] for waveform in raw_speech
]
if isinstance(audio_features[0] , _UpperCAmelCase ):
UpperCamelCase_ = [np.asarray(_UpperCAmelCase , dtype=np.floataa ) for feature in audio_features]
# Create audio attention mask
UpperCamelCase_ = max(
[ceil(feature.shape[0] / self.patch_size[0] ) * self.freq_len for feature in audio_features] ) # The maximum number of audio patches in a batch
if return_attention_mask:
UpperCamelCase_ = [
(ceil(feature.shape[0] / self.patch_size[0] ) * self.freq_len) * [1]
+ (max_patch_len - ceil(feature.shape[0] / self.patch_size[0] ) * self.freq_len) * [0]
for feature in audio_features
]
UpperCamelCase_ = np.array(_UpperCAmelCase ).astype(np.floataa )
# convert into correct format for padding
UpperCamelCase_ = max_patch_len // self.freq_len * self.patch_size[0] # The maximum audio size in a batch
UpperCamelCase_ = np.ones([len(_UpperCAmelCase ), 1, max_time_len, self.feature_size] ).astype(np.floataa )
UpperCamelCase_ = padded_audio_features * self.padding_value
for i in range(len(_UpperCAmelCase ) ):
UpperCamelCase_ = audio_features[i]
UpperCamelCase_ = feature
# return as BatchFeature
if return_attention_mask:
UpperCamelCase_ = {'audio_values': padded_audio_features, 'audio_mask': audio_mask}
else:
UpperCamelCase_ = {'audio_values': padded_audio_features}
UpperCamelCase_ = BatchFeature(data=_UpperCAmelCase , tensor_type=_UpperCAmelCase )
return encoded_inputs
| 23 |
def _snake_case (__lowercase):
UpperCamelCase_ = 1
for i in range(1 , num + 1):
fact *= i
return fact
def _snake_case (__lowercase):
UpperCamelCase_ = 0
while number > 0:
UpperCamelCase_ = number % 10
sum_of_digits += last_digit
UpperCamelCase_ = number // 10 # Removing the last_digit from the given number
return sum_of_digits
def _snake_case (__lowercase = 100):
UpperCamelCase_ = factorial(__lowercase)
UpperCamelCase_ = split_and_add(__lowercase)
return result
if __name__ == "__main__":
print(solution(int(input("""Enter the Number: """).strip())))
| 23 | 1 |
from typing import Dict, List, Optional, Union
import numpy as np
from ...image_processing_utils import BaseImageProcessor, BatchFeature, get_size_dict
from ...image_transforms import (
center_crop,
get_resize_output_image_size,
normalize,
rescale,
resize,
to_channel_dimension_format,
)
from ...image_utils import (
IMAGENET_STANDARD_MEAN,
IMAGENET_STANDARD_STD,
ChannelDimension,
ImageInput,
PILImageResampling,
is_valid_image,
to_numpy_array,
valid_images,
)
from ...utils import TensorType, is_vision_available, logging
if is_vision_available():
import PIL
snake_case__ : str = logging.get_logger(__name__)
def _snake_case (__lowercase):
if isinstance(__lowercase , (list, tuple)) and isinstance(videos[0] , (list, tuple)) and is_valid_image(videos[0][0]):
return videos
elif isinstance(__lowercase , (list, tuple)) and is_valid_image(videos[0]):
return [videos]
elif is_valid_image(__lowercase):
return [[videos]]
raise ValueError(f"""Could not make batched video from {videos}""")
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
A_ = ["""pixel_values"""]
def __init__( self , _UpperCAmelCase = True , _UpperCAmelCase = None , _UpperCAmelCase = PILImageResampling.BILINEAR , _UpperCAmelCase = True , _UpperCAmelCase = None , _UpperCAmelCase = True , _UpperCAmelCase = 1 / 255 , _UpperCAmelCase = True , _UpperCAmelCase = None , _UpperCAmelCase = None , **_UpperCAmelCase , ) -> None:
super().__init__(**_UpperCAmelCase )
UpperCamelCase_ = size if size is not None else {'shortest_edge': 224}
UpperCamelCase_ = get_size_dict(_UpperCAmelCase , default_to_square=_UpperCAmelCase )
UpperCamelCase_ = crop_size if crop_size is not None else {'height': 224, 'width': 224}
UpperCamelCase_ = get_size_dict(_UpperCAmelCase , param_name='crop_size' )
UpperCamelCase_ = do_resize
UpperCamelCase_ = size
UpperCamelCase_ = do_center_crop
UpperCamelCase_ = crop_size
UpperCamelCase_ = resample
UpperCamelCase_ = do_rescale
UpperCamelCase_ = rescale_factor
UpperCamelCase_ = do_normalize
UpperCamelCase_ = image_mean if image_mean is not None else IMAGENET_STANDARD_MEAN
UpperCamelCase_ = image_std if image_std is not None else IMAGENET_STANDARD_STD
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase = PILImageResampling.BILINEAR , _UpperCAmelCase = None , **_UpperCAmelCase , ) -> np.ndarray:
UpperCamelCase_ = get_size_dict(_UpperCAmelCase , default_to_square=_UpperCAmelCase )
if "shortest_edge" in size:
UpperCamelCase_ = get_resize_output_image_size(_UpperCAmelCase , size['shortest_edge'] , default_to_square=_UpperCAmelCase )
elif "height" in size and "width" in size:
UpperCamelCase_ = (size['height'], size['width'])
else:
raise ValueError(f"""Size must have 'height' and 'width' or 'shortest_edge' as keys. Got {size.keys()}""" )
return resize(_UpperCAmelCase , size=_UpperCAmelCase , resample=_UpperCAmelCase , data_format=_UpperCAmelCase , **_UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase = None , **_UpperCAmelCase , ) -> np.ndarray:
UpperCamelCase_ = get_size_dict(_UpperCAmelCase )
if "height" not in size or "width" not in size:
raise ValueError(f"""Size must have 'height' and 'width' as keys. Got {size.keys()}""" )
return center_crop(_UpperCAmelCase , size=(size['height'], size['width']) , data_format=_UpperCAmelCase , **_UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase = None , **_UpperCAmelCase , ) -> int:
return rescale(_UpperCAmelCase , scale=_UpperCAmelCase , data_format=_UpperCAmelCase , **_UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase = None , **_UpperCAmelCase , ) -> np.ndarray:
return normalize(_UpperCAmelCase , mean=_UpperCAmelCase , std=_UpperCAmelCase , data_format=_UpperCAmelCase , **_UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = ChannelDimension.FIRST , ) -> np.ndarray:
if do_resize and size is None or resample is None:
raise ValueError('Size and resample must be specified if do_resize is True.' )
if do_center_crop and crop_size is None:
raise ValueError('Crop size must be specified if do_center_crop is True.' )
if do_rescale and rescale_factor is None:
raise ValueError('Rescale factor must be specified if do_rescale is True.' )
if do_normalize and (image_mean is None or image_std is None):
raise ValueError('Image mean and std must be specified if do_normalize is True.' )
# All transformations expect numpy arrays.
UpperCamelCase_ = to_numpy_array(_UpperCAmelCase )
if do_resize:
UpperCamelCase_ = self.resize(image=_UpperCAmelCase , size=_UpperCAmelCase , resample=_UpperCAmelCase )
if do_center_crop:
UpperCamelCase_ = self.center_crop(_UpperCAmelCase , size=_UpperCAmelCase )
if do_rescale:
UpperCamelCase_ = self.rescale(image=_UpperCAmelCase , scale=_UpperCAmelCase )
if do_normalize:
UpperCamelCase_ = self.normalize(image=_UpperCAmelCase , mean=_UpperCAmelCase , std=_UpperCAmelCase )
UpperCamelCase_ = to_channel_dimension_format(_UpperCAmelCase , _UpperCAmelCase )
return image
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = ChannelDimension.FIRST , **_UpperCAmelCase , ) -> PIL.Image.Image:
UpperCamelCase_ = do_resize if do_resize is not None else self.do_resize
UpperCamelCase_ = resample if resample is not None else self.resample
UpperCamelCase_ = do_center_crop if do_center_crop is not None else self.do_center_crop
UpperCamelCase_ = do_rescale if do_rescale is not None else self.do_rescale
UpperCamelCase_ = rescale_factor if rescale_factor is not None else self.rescale_factor
UpperCamelCase_ = do_normalize if do_normalize is not None else self.do_normalize
UpperCamelCase_ = image_mean if image_mean is not None else self.image_mean
UpperCamelCase_ = image_std if image_std is not None else self.image_std
UpperCamelCase_ = size if size is not None else self.size
UpperCamelCase_ = get_size_dict(_UpperCAmelCase , default_to_square=_UpperCAmelCase )
UpperCamelCase_ = crop_size if crop_size is not None else self.crop_size
UpperCamelCase_ = get_size_dict(_UpperCAmelCase , param_name='crop_size' )
if not valid_images(_UpperCAmelCase ):
raise ValueError(
'Invalid image type. Must be of type PIL.Image.Image, numpy.ndarray, '
'torch.Tensor, tf.Tensor or jax.ndarray.' )
UpperCamelCase_ = make_batched(_UpperCAmelCase )
UpperCamelCase_ = [
[
self._preprocess_image(
image=_UpperCAmelCase , do_resize=_UpperCAmelCase , size=_UpperCAmelCase , resample=_UpperCAmelCase , do_center_crop=_UpperCAmelCase , crop_size=_UpperCAmelCase , do_rescale=_UpperCAmelCase , rescale_factor=_UpperCAmelCase , do_normalize=_UpperCAmelCase , image_mean=_UpperCAmelCase , image_std=_UpperCAmelCase , data_format=_UpperCAmelCase , )
for img in video
]
for video in videos
]
UpperCamelCase_ = {'pixel_values': videos}
return BatchFeature(data=_UpperCAmelCase , tensor_type=_UpperCAmelCase )
| 23 |
from typing import Dict, List, Optional, Union
import numpy as np
from ...image_processing_utils import BaseImageProcessor, BatchFeature, get_size_dict
from ...image_transforms import (
center_crop,
get_resize_output_image_size,
normalize,
rescale,
resize,
to_channel_dimension_format,
)
from ...image_utils import (
IMAGENET_STANDARD_MEAN,
IMAGENET_STANDARD_STD,
ChannelDimension,
ImageInput,
PILImageResampling,
is_valid_image,
to_numpy_array,
valid_images,
)
from ...utils import TensorType, is_vision_available, logging
if is_vision_available():
import PIL
snake_case__ : str = logging.get_logger(__name__)
def _snake_case (__lowercase):
if isinstance(__lowercase , (list, tuple)) and isinstance(videos[0] , (list, tuple)) and is_valid_image(videos[0][0]):
return videos
elif isinstance(__lowercase , (list, tuple)) and is_valid_image(videos[0]):
return [videos]
elif is_valid_image(__lowercase):
return [[videos]]
raise ValueError(f"""Could not make batched video from {videos}""")
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
A_ = ["""pixel_values"""]
def __init__( self , _UpperCAmelCase = True , _UpperCAmelCase = None , _UpperCAmelCase = PILImageResampling.BILINEAR , _UpperCAmelCase = True , _UpperCAmelCase = None , _UpperCAmelCase = True , _UpperCAmelCase = 1 / 255 , _UpperCAmelCase = True , _UpperCAmelCase = None , _UpperCAmelCase = None , **_UpperCAmelCase , ) -> None:
super().__init__(**_UpperCAmelCase )
UpperCamelCase_ = size if size is not None else {'shortest_edge': 224}
UpperCamelCase_ = get_size_dict(_UpperCAmelCase , default_to_square=_UpperCAmelCase )
UpperCamelCase_ = crop_size if crop_size is not None else {'height': 224, 'width': 224}
UpperCamelCase_ = get_size_dict(_UpperCAmelCase , param_name='crop_size' )
UpperCamelCase_ = do_resize
UpperCamelCase_ = size
UpperCamelCase_ = do_center_crop
UpperCamelCase_ = crop_size
UpperCamelCase_ = resample
UpperCamelCase_ = do_rescale
UpperCamelCase_ = rescale_factor
UpperCamelCase_ = do_normalize
UpperCamelCase_ = image_mean if image_mean is not None else IMAGENET_STANDARD_MEAN
UpperCamelCase_ = image_std if image_std is not None else IMAGENET_STANDARD_STD
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase = PILImageResampling.BILINEAR , _UpperCAmelCase = None , **_UpperCAmelCase , ) -> np.ndarray:
UpperCamelCase_ = get_size_dict(_UpperCAmelCase , default_to_square=_UpperCAmelCase )
if "shortest_edge" in size:
UpperCamelCase_ = get_resize_output_image_size(_UpperCAmelCase , size['shortest_edge'] , default_to_square=_UpperCAmelCase )
elif "height" in size and "width" in size:
UpperCamelCase_ = (size['height'], size['width'])
else:
raise ValueError(f"""Size must have 'height' and 'width' or 'shortest_edge' as keys. Got {size.keys()}""" )
return resize(_UpperCAmelCase , size=_UpperCAmelCase , resample=_UpperCAmelCase , data_format=_UpperCAmelCase , **_UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase = None , **_UpperCAmelCase , ) -> np.ndarray:
UpperCamelCase_ = get_size_dict(_UpperCAmelCase )
if "height" not in size or "width" not in size:
raise ValueError(f"""Size must have 'height' and 'width' as keys. Got {size.keys()}""" )
return center_crop(_UpperCAmelCase , size=(size['height'], size['width']) , data_format=_UpperCAmelCase , **_UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase = None , **_UpperCAmelCase , ) -> int:
return rescale(_UpperCAmelCase , scale=_UpperCAmelCase , data_format=_UpperCAmelCase , **_UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase = None , **_UpperCAmelCase , ) -> np.ndarray:
return normalize(_UpperCAmelCase , mean=_UpperCAmelCase , std=_UpperCAmelCase , data_format=_UpperCAmelCase , **_UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = ChannelDimension.FIRST , ) -> np.ndarray:
if do_resize and size is None or resample is None:
raise ValueError('Size and resample must be specified if do_resize is True.' )
if do_center_crop and crop_size is None:
raise ValueError('Crop size must be specified if do_center_crop is True.' )
if do_rescale and rescale_factor is None:
raise ValueError('Rescale factor must be specified if do_rescale is True.' )
if do_normalize and (image_mean is None or image_std is None):
raise ValueError('Image mean and std must be specified if do_normalize is True.' )
# All transformations expect numpy arrays.
UpperCamelCase_ = to_numpy_array(_UpperCAmelCase )
if do_resize:
UpperCamelCase_ = self.resize(image=_UpperCAmelCase , size=_UpperCAmelCase , resample=_UpperCAmelCase )
if do_center_crop:
UpperCamelCase_ = self.center_crop(_UpperCAmelCase , size=_UpperCAmelCase )
if do_rescale:
UpperCamelCase_ = self.rescale(image=_UpperCAmelCase , scale=_UpperCAmelCase )
if do_normalize:
UpperCamelCase_ = self.normalize(image=_UpperCAmelCase , mean=_UpperCAmelCase , std=_UpperCAmelCase )
UpperCamelCase_ = to_channel_dimension_format(_UpperCAmelCase , _UpperCAmelCase )
return image
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = ChannelDimension.FIRST , **_UpperCAmelCase , ) -> PIL.Image.Image:
UpperCamelCase_ = do_resize if do_resize is not None else self.do_resize
UpperCamelCase_ = resample if resample is not None else self.resample
UpperCamelCase_ = do_center_crop if do_center_crop is not None else self.do_center_crop
UpperCamelCase_ = do_rescale if do_rescale is not None else self.do_rescale
UpperCamelCase_ = rescale_factor if rescale_factor is not None else self.rescale_factor
UpperCamelCase_ = do_normalize if do_normalize is not None else self.do_normalize
UpperCamelCase_ = image_mean if image_mean is not None else self.image_mean
UpperCamelCase_ = image_std if image_std is not None else self.image_std
UpperCamelCase_ = size if size is not None else self.size
UpperCamelCase_ = get_size_dict(_UpperCAmelCase , default_to_square=_UpperCAmelCase )
UpperCamelCase_ = crop_size if crop_size is not None else self.crop_size
UpperCamelCase_ = get_size_dict(_UpperCAmelCase , param_name='crop_size' )
if not valid_images(_UpperCAmelCase ):
raise ValueError(
'Invalid image type. Must be of type PIL.Image.Image, numpy.ndarray, '
'torch.Tensor, tf.Tensor or jax.ndarray.' )
UpperCamelCase_ = make_batched(_UpperCAmelCase )
UpperCamelCase_ = [
[
self._preprocess_image(
image=_UpperCAmelCase , do_resize=_UpperCAmelCase , size=_UpperCAmelCase , resample=_UpperCAmelCase , do_center_crop=_UpperCAmelCase , crop_size=_UpperCAmelCase , do_rescale=_UpperCAmelCase , rescale_factor=_UpperCAmelCase , do_normalize=_UpperCAmelCase , image_mean=_UpperCAmelCase , image_std=_UpperCAmelCase , data_format=_UpperCAmelCase , )
for img in video
]
for video in videos
]
UpperCamelCase_ = {'pixel_values': videos}
return BatchFeature(data=_UpperCAmelCase , tensor_type=_UpperCAmelCase )
| 23 | 1 |
import unittest
import numpy as np
from transformers.testing_utils import require_flax, require_tf, require_torch
from transformers.utils import (
expand_dims,
flatten_dict,
is_flax_available,
is_tf_available,
is_torch_available,
reshape,
squeeze,
transpose,
)
if is_flax_available():
import jax.numpy as jnp
if is_tf_available():
import tensorflow as tf
if is_torch_available():
import torch
class _a ( unittest.TestCase ):
"""simple docstring"""
def _UpperCAmelCase ( self ) -> str:
UpperCamelCase_ = {
'task_specific_params': {
'summarization': {'length_penalty': 1.0, 'max_length': 128, 'min_length': 12, 'num_beams': 4},
'summarization_cnn': {'length_penalty': 2.0, 'max_length': 142, 'min_length': 56, 'num_beams': 4},
'summarization_xsum': {'length_penalty': 1.0, 'max_length': 62, 'min_length': 11, 'num_beams': 6},
}
}
UpperCamelCase_ = {
'task_specific_params.summarization.length_penalty': 1.0,
'task_specific_params.summarization.max_length': 128,
'task_specific_params.summarization.min_length': 12,
'task_specific_params.summarization.num_beams': 4,
'task_specific_params.summarization_cnn.length_penalty': 2.0,
'task_specific_params.summarization_cnn.max_length': 142,
'task_specific_params.summarization_cnn.min_length': 56,
'task_specific_params.summarization_cnn.num_beams': 4,
'task_specific_params.summarization_xsum.length_penalty': 1.0,
'task_specific_params.summarization_xsum.max_length': 62,
'task_specific_params.summarization_xsum.min_length': 11,
'task_specific_params.summarization_xsum.num_beams': 6,
}
self.assertEqual(flatten_dict(_UpperCAmelCase ) , _UpperCAmelCase )
def _UpperCAmelCase ( self ) -> Dict:
UpperCamelCase_ = np.random.randn(3 , 4 )
self.assertTrue(np.allclose(transpose(_UpperCAmelCase ) , x.transpose() ) )
UpperCamelCase_ = np.random.randn(3 , 4 , 5 )
self.assertTrue(np.allclose(transpose(_UpperCAmelCase , axes=(1, 2, 0) ) , x.transpose((1, 2, 0) ) ) )
@require_torch
def _UpperCAmelCase ( self ) -> int:
UpperCamelCase_ = np.random.randn(3 , 4 )
UpperCamelCase_ = torch.tensor(_UpperCAmelCase )
self.assertTrue(np.allclose(transpose(_UpperCAmelCase ) , transpose(_UpperCAmelCase ).numpy() ) )
UpperCamelCase_ = np.random.randn(3 , 4 , 5 )
UpperCamelCase_ = torch.tensor(_UpperCAmelCase )
self.assertTrue(np.allclose(transpose(_UpperCAmelCase , axes=(1, 2, 0) ) , transpose(_UpperCAmelCase , axes=(1, 2, 0) ).numpy() ) )
@require_tf
def _UpperCAmelCase ( self ) -> Optional[int]:
UpperCamelCase_ = np.random.randn(3 , 4 )
UpperCamelCase_ = tf.constant(_UpperCAmelCase )
self.assertTrue(np.allclose(transpose(_UpperCAmelCase ) , transpose(_UpperCAmelCase ).numpy() ) )
UpperCamelCase_ = np.random.randn(3 , 4 , 5 )
UpperCamelCase_ = tf.constant(_UpperCAmelCase )
self.assertTrue(np.allclose(transpose(_UpperCAmelCase , axes=(1, 2, 0) ) , transpose(_UpperCAmelCase , axes=(1, 2, 0) ).numpy() ) )
@require_flax
def _UpperCAmelCase ( self ) -> Any:
UpperCamelCase_ = np.random.randn(3 , 4 )
UpperCamelCase_ = jnp.array(_UpperCAmelCase )
self.assertTrue(np.allclose(transpose(_UpperCAmelCase ) , np.asarray(transpose(_UpperCAmelCase ) ) ) )
UpperCamelCase_ = np.random.randn(3 , 4 , 5 )
UpperCamelCase_ = jnp.array(_UpperCAmelCase )
self.assertTrue(np.allclose(transpose(_UpperCAmelCase , axes=(1, 2, 0) ) , np.asarray(transpose(_UpperCAmelCase , axes=(1, 2, 0) ) ) ) )
def _UpperCAmelCase ( self ) -> Tuple:
UpperCamelCase_ = np.random.randn(3 , 4 )
self.assertTrue(np.allclose(reshape(_UpperCAmelCase , (4, 3) ) , np.reshape(_UpperCAmelCase , (4, 3) ) ) )
UpperCamelCase_ = np.random.randn(3 , 4 , 5 )
self.assertTrue(np.allclose(reshape(_UpperCAmelCase , (12, 5) ) , np.reshape(_UpperCAmelCase , (12, 5) ) ) )
@require_torch
def _UpperCAmelCase ( self ) -> List[str]:
UpperCamelCase_ = np.random.randn(3 , 4 )
UpperCamelCase_ = torch.tensor(_UpperCAmelCase )
self.assertTrue(np.allclose(reshape(_UpperCAmelCase , (4, 3) ) , reshape(_UpperCAmelCase , (4, 3) ).numpy() ) )
UpperCamelCase_ = np.random.randn(3 , 4 , 5 )
UpperCamelCase_ = torch.tensor(_UpperCAmelCase )
self.assertTrue(np.allclose(reshape(_UpperCAmelCase , (12, 5) ) , reshape(_UpperCAmelCase , (12, 5) ).numpy() ) )
@require_tf
def _UpperCAmelCase ( self ) -> List[Any]:
UpperCamelCase_ = np.random.randn(3 , 4 )
UpperCamelCase_ = tf.constant(_UpperCAmelCase )
self.assertTrue(np.allclose(reshape(_UpperCAmelCase , (4, 3) ) , reshape(_UpperCAmelCase , (4, 3) ).numpy() ) )
UpperCamelCase_ = np.random.randn(3 , 4 , 5 )
UpperCamelCase_ = tf.constant(_UpperCAmelCase )
self.assertTrue(np.allclose(reshape(_UpperCAmelCase , (12, 5) ) , reshape(_UpperCAmelCase , (12, 5) ).numpy() ) )
@require_flax
def _UpperCAmelCase ( self ) -> List[str]:
UpperCamelCase_ = np.random.randn(3 , 4 )
UpperCamelCase_ = jnp.array(_UpperCAmelCase )
self.assertTrue(np.allclose(reshape(_UpperCAmelCase , (4, 3) ) , np.asarray(reshape(_UpperCAmelCase , (4, 3) ) ) ) )
UpperCamelCase_ = np.random.randn(3 , 4 , 5 )
UpperCamelCase_ = jnp.array(_UpperCAmelCase )
self.assertTrue(np.allclose(reshape(_UpperCAmelCase , (12, 5) ) , np.asarray(reshape(_UpperCAmelCase , (12, 5) ) ) ) )
def _UpperCAmelCase ( self ) -> int:
UpperCamelCase_ = np.random.randn(1 , 3 , 4 )
self.assertTrue(np.allclose(squeeze(_UpperCAmelCase ) , np.squeeze(_UpperCAmelCase ) ) )
UpperCamelCase_ = np.random.randn(1 , 4 , 1 , 5 )
self.assertTrue(np.allclose(squeeze(_UpperCAmelCase , axis=2 ) , np.squeeze(_UpperCAmelCase , axis=2 ) ) )
@require_torch
def _UpperCAmelCase ( self ) -> Union[str, Any]:
UpperCamelCase_ = np.random.randn(1 , 3 , 4 )
UpperCamelCase_ = torch.tensor(_UpperCAmelCase )
self.assertTrue(np.allclose(squeeze(_UpperCAmelCase ) , squeeze(_UpperCAmelCase ).numpy() ) )
UpperCamelCase_ = np.random.randn(1 , 4 , 1 , 5 )
UpperCamelCase_ = torch.tensor(_UpperCAmelCase )
self.assertTrue(np.allclose(squeeze(_UpperCAmelCase , axis=2 ) , squeeze(_UpperCAmelCase , axis=2 ).numpy() ) )
@require_tf
def _UpperCAmelCase ( self ) -> Dict:
UpperCamelCase_ = np.random.randn(1 , 3 , 4 )
UpperCamelCase_ = tf.constant(_UpperCAmelCase )
self.assertTrue(np.allclose(squeeze(_UpperCAmelCase ) , squeeze(_UpperCAmelCase ).numpy() ) )
UpperCamelCase_ = np.random.randn(1 , 4 , 1 , 5 )
UpperCamelCase_ = tf.constant(_UpperCAmelCase )
self.assertTrue(np.allclose(squeeze(_UpperCAmelCase , axis=2 ) , squeeze(_UpperCAmelCase , axis=2 ).numpy() ) )
@require_flax
def _UpperCAmelCase ( self ) -> int:
UpperCamelCase_ = np.random.randn(1 , 3 , 4 )
UpperCamelCase_ = jnp.array(_UpperCAmelCase )
self.assertTrue(np.allclose(squeeze(_UpperCAmelCase ) , np.asarray(squeeze(_UpperCAmelCase ) ) ) )
UpperCamelCase_ = np.random.randn(1 , 4 , 1 , 5 )
UpperCamelCase_ = jnp.array(_UpperCAmelCase )
self.assertTrue(np.allclose(squeeze(_UpperCAmelCase , axis=2 ) , np.asarray(squeeze(_UpperCAmelCase , axis=2 ) ) ) )
def _UpperCAmelCase ( self ) -> Union[str, Any]:
UpperCamelCase_ = np.random.randn(3 , 4 )
self.assertTrue(np.allclose(expand_dims(_UpperCAmelCase , axis=1 ) , np.expand_dims(_UpperCAmelCase , axis=1 ) ) )
@require_torch
def _UpperCAmelCase ( self ) -> List[str]:
UpperCamelCase_ = np.random.randn(3 , 4 )
UpperCamelCase_ = torch.tensor(_UpperCAmelCase )
self.assertTrue(np.allclose(expand_dims(_UpperCAmelCase , axis=1 ) , expand_dims(_UpperCAmelCase , axis=1 ).numpy() ) )
@require_tf
def _UpperCAmelCase ( self ) -> Optional[int]:
UpperCamelCase_ = np.random.randn(3 , 4 )
UpperCamelCase_ = tf.constant(_UpperCAmelCase )
self.assertTrue(np.allclose(expand_dims(_UpperCAmelCase , axis=1 ) , expand_dims(_UpperCAmelCase , axis=1 ).numpy() ) )
@require_flax
def _UpperCAmelCase ( self ) -> Any:
UpperCamelCase_ = np.random.randn(3 , 4 )
UpperCamelCase_ = jnp.array(_UpperCAmelCase )
self.assertTrue(np.allclose(expand_dims(_UpperCAmelCase , axis=1 ) , np.asarray(expand_dims(_UpperCAmelCase , axis=1 ) ) ) )
| 23 |
# DISCLAIMER: This file is strongly influenced by https://github.com/yang-song/score_sde_pytorch
import math
from dataclasses import dataclass
from typing import Optional, Tuple, Union
import torch
from ..configuration_utils import ConfigMixin, register_to_config
from ..utils import BaseOutput, randn_tensor
from .scheduling_utils import SchedulerMixin, SchedulerOutput
@dataclass
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
A_ = 42
A_ = 42
class _a ( UpperCAmelCase__ , UpperCAmelCase__ ):
"""simple docstring"""
A_ = 1
@register_to_config
def __init__( self , _UpperCAmelCase = 2000 , _UpperCAmelCase = 0.1_5 , _UpperCAmelCase = 0.0_1 , _UpperCAmelCase = 1_3_4_8.0 , _UpperCAmelCase = 1e-5 , _UpperCAmelCase = 1 , ) -> Tuple:
# standard deviation of the initial noise distribution
UpperCamelCase_ = sigma_max
# setable values
UpperCamelCase_ = None
self.set_sigmas(_UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase = None ) -> torch.FloatTensor:
return sample
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase = None , _UpperCAmelCase = None ) -> str:
UpperCamelCase_ = sampling_eps if sampling_eps is not None else self.config.sampling_eps
UpperCamelCase_ = torch.linspace(1 , _UpperCAmelCase , _UpperCAmelCase , device=_UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None ) -> Any:
UpperCamelCase_ = sigma_min if sigma_min is not None else self.config.sigma_min
UpperCamelCase_ = sigma_max if sigma_max is not None else self.config.sigma_max
UpperCamelCase_ = sampling_eps if sampling_eps is not None else self.config.sampling_eps
if self.timesteps is None:
self.set_timesteps(_UpperCAmelCase , _UpperCAmelCase )
UpperCamelCase_ = sigma_min * (sigma_max / sigma_min) ** (self.timesteps / sampling_eps)
UpperCamelCase_ = torch.exp(torch.linspace(math.log(_UpperCAmelCase ) , math.log(_UpperCAmelCase ) , _UpperCAmelCase ) )
UpperCamelCase_ = torch.tensor([sigma_min * (sigma_max / sigma_min) ** t for t in self.timesteps] )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> List[Any]:
return torch.where(
timesteps == 0 , torch.zeros_like(t.to(timesteps.device ) ) , self.discrete_sigmas[timesteps - 1].to(timesteps.device ) , )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase = None , _UpperCAmelCase = True , ) -> Union[SdeVeOutput, Tuple]:
if self.timesteps is None:
raise ValueError(
'`self.timesteps` is not set, you need to run \'set_timesteps\' after creating the scheduler' )
UpperCamelCase_ = timestep * torch.ones(
sample.shape[0] , device=sample.device ) # torch.repeat_interleave(timestep, sample.shape[0])
UpperCamelCase_ = (timestep * (len(self.timesteps ) - 1)).long()
# mps requires indices to be in the same device, so we use cpu as is the default with cuda
UpperCamelCase_ = timesteps.to(self.discrete_sigmas.device )
UpperCamelCase_ = self.discrete_sigmas[timesteps].to(sample.device )
UpperCamelCase_ = self.get_adjacent_sigma(_UpperCAmelCase , _UpperCAmelCase ).to(sample.device )
UpperCamelCase_ = torch.zeros_like(_UpperCAmelCase )
UpperCamelCase_ = (sigma**2 - adjacent_sigma**2) ** 0.5
# equation 6 in the paper: the model_output modeled by the network is grad_x log pt(x)
# also equation 47 shows the analog from SDE models to ancestral sampling methods
UpperCamelCase_ = diffusion.flatten()
while len(diffusion.shape ) < len(sample.shape ):
UpperCamelCase_ = diffusion.unsqueeze(-1 )
UpperCamelCase_ = drift - diffusion**2 * model_output
# equation 6: sample noise for the diffusion term of
UpperCamelCase_ = randn_tensor(
sample.shape , layout=sample.layout , generator=_UpperCAmelCase , device=sample.device , dtype=sample.dtype )
UpperCamelCase_ = sample - drift # subtract because `dt` is a small negative timestep
# TODO is the variable diffusion the correct scaling term for the noise?
UpperCamelCase_ = prev_sample_mean + diffusion * noise # add impact of diffusion field g
if not return_dict:
return (prev_sample, prev_sample_mean)
return SdeVeOutput(prev_sample=_UpperCAmelCase , prev_sample_mean=_UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase = None , _UpperCAmelCase = True , ) -> Union[SchedulerOutput, Tuple]:
if self.timesteps is None:
raise ValueError(
'`self.timesteps` is not set, you need to run \'set_timesteps\' after creating the scheduler' )
# For small batch sizes, the paper "suggest replacing norm(z) with sqrt(d), where d is the dim. of z"
# sample noise for correction
UpperCamelCase_ = randn_tensor(sample.shape , layout=sample.layout , generator=_UpperCAmelCase ).to(sample.device )
# compute step size from the model_output, the noise, and the snr
UpperCamelCase_ = torch.norm(model_output.reshape(model_output.shape[0] , -1 ) , dim=-1 ).mean()
UpperCamelCase_ = torch.norm(noise.reshape(noise.shape[0] , -1 ) , dim=-1 ).mean()
UpperCamelCase_ = (self.config.snr * noise_norm / grad_norm) ** 2 * 2
UpperCamelCase_ = step_size * torch.ones(sample.shape[0] ).to(sample.device )
# self.repeat_scalar(step_size, sample.shape[0])
# compute corrected sample: model_output term and noise term
UpperCamelCase_ = step_size.flatten()
while len(step_size.shape ) < len(sample.shape ):
UpperCamelCase_ = step_size.unsqueeze(-1 )
UpperCamelCase_ = sample + step_size * model_output
UpperCamelCase_ = prev_sample_mean + ((step_size * 2) ** 0.5) * noise
if not return_dict:
return (prev_sample,)
return SchedulerOutput(prev_sample=_UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , ) -> torch.FloatTensor:
# Make sure sigmas and timesteps have the same device and dtype as original_samples
UpperCamelCase_ = timesteps.to(original_samples.device )
UpperCamelCase_ = self.discrete_sigmas.to(original_samples.device )[timesteps]
UpperCamelCase_ = (
noise * sigmas[:, None, None, None]
if noise is not None
else torch.randn_like(_UpperCAmelCase ) * sigmas[:, None, None, None]
)
UpperCamelCase_ = noise + original_samples
return noisy_samples
def __len__( self ) -> Optional[int]:
return self.config.num_train_timesteps
| 23 | 1 |
import argparse
import os
import evaluate
import torch
from datasets import load_dataset
from torch.optim import AdamW
from torch.utils.data import DataLoader
from transformers import AutoModelForSequenceClassification, AutoTokenizer, get_linear_schedule_with_warmup, set_seed
from accelerate import Accelerator, DistributedType
########################################################################
# This is a fully working simple example to use Accelerate
# and perform gradient accumulation
#
# This example trains a Bert base model on GLUE MRPC
# in any of the following settings (with the same script):
# - single CPU or single GPU
# - multi GPUS (using PyTorch distributed mode)
# - (multi) TPUs
# - fp16 (mixed-precision) or fp32 (normal precision)
#
# To run it in each of these various modes, follow the instructions
# in the readme for examples:
# https://github.com/huggingface/accelerate/tree/main/examples
#
########################################################################
snake_case__ : Dict = 1_6
snake_case__ : List[str] = 3_2
def _snake_case (__lowercase , __lowercase = 16):
UpperCamelCase_ = AutoTokenizer.from_pretrained('bert-base-cased')
UpperCamelCase_ = load_dataset('glue' , 'mrpc')
def tokenize_function(__lowercase):
# max_length=None => use the model max length (it's actually the default)
UpperCamelCase_ = tokenizer(examples['sentence1'] , examples['sentence2'] , truncation=__lowercase , max_length=__lowercase)
return outputs
# Apply the method we just defined to all the examples in all the splits of the dataset
# starting with the main process first:
with accelerator.main_process_first():
UpperCamelCase_ = datasets.map(
__lowercase , batched=__lowercase , remove_columns=['idx', 'sentence1', 'sentence2'] , )
# We also rename the 'label' column to 'labels' which is the expected name for labels by the models of the
# transformers library
UpperCamelCase_ = tokenized_datasets.rename_column('label' , 'labels')
def collate_fn(__lowercase):
# On TPU it's best to pad everything to the same length or training will be very slow.
UpperCamelCase_ = 128 if accelerator.distributed_type == DistributedType.TPU else None
# When using mixed precision we want round multiples of 8/16
if accelerator.mixed_precision == "fp8":
UpperCamelCase_ = 16
elif accelerator.mixed_precision != "no":
UpperCamelCase_ = 8
else:
UpperCamelCase_ = None
return tokenizer.pad(
__lowercase , padding='longest' , max_length=__lowercase , pad_to_multiple_of=__lowercase , return_tensors='pt' , )
# Instantiate dataloaders.
UpperCamelCase_ = DataLoader(
tokenized_datasets['train'] , shuffle=__lowercase , collate_fn=__lowercase , batch_size=__lowercase)
UpperCamelCase_ = DataLoader(
tokenized_datasets['validation'] , shuffle=__lowercase , collate_fn=__lowercase , batch_size=__lowercase)
return train_dataloader, eval_dataloader
# For testing only
if os.environ.get("""TESTING_MOCKED_DATALOADERS""", None) == "1":
from accelerate.test_utils.training import mocked_dataloaders
snake_case__ : List[str] = mocked_dataloaders # noqa: F811
def _snake_case (__lowercase , __lowercase):
# For testing only
if os.environ.get('TESTING_MOCKED_DATALOADERS' , __lowercase) == "1":
UpperCamelCase_ = 2
# New Code #
UpperCamelCase_ = int(args.gradient_accumulation_steps)
# Initialize accelerator
UpperCamelCase_ = Accelerator(
cpu=args.cpu , mixed_precision=args.mixed_precision , gradient_accumulation_steps=__lowercase)
if accelerator.distributed_type == DistributedType.TPU and gradient_accumulation_steps > 1:
raise NotImplementedError(
'Gradient accumulation on TPUs is currently not supported. Pass `gradient_accumulation_steps=1`')
# Sample hyper-parameters for learning rate, batch size, seed and a few other HPs
UpperCamelCase_ = config['lr']
UpperCamelCase_ = int(config['num_epochs'])
UpperCamelCase_ = int(config['seed'])
UpperCamelCase_ = int(config['batch_size'])
UpperCamelCase_ = evaluate.load('glue' , 'mrpc')
set_seed(__lowercase)
UpperCamelCase_ , UpperCamelCase_ = get_dataloaders(__lowercase , __lowercase)
# Instantiate the model (we build the model here so that the seed also control new weights initialization)
UpperCamelCase_ = AutoModelForSequenceClassification.from_pretrained('bert-base-cased' , return_dict=__lowercase)
# We could avoid this line since the accelerator is set with `device_placement=True` (default value).
# Note that if you are placing tensors on devices manually, this line absolutely needs to be before the optimizer
# creation otherwise training will not work on TPU (`accelerate` will kindly throw an error to make us aware of that).
UpperCamelCase_ = model.to(accelerator.device)
# Instantiate optimizer
UpperCamelCase_ = AdamW(params=model.parameters() , lr=__lowercase)
# Instantiate scheduler
UpperCamelCase_ = get_linear_schedule_with_warmup(
optimizer=__lowercase , num_warmup_steps=100 , num_training_steps=(len(__lowercase) * num_epochs) , )
# Prepare everything
# There is no specific order to remember, we just need to unpack the objects in the same order we gave them to the
# prepare method.
UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ = accelerator.prepare(
__lowercase , __lowercase , __lowercase , __lowercase , __lowercase)
# Now we train the model
for epoch in range(__lowercase):
model.train()
for step, batch in enumerate(__lowercase):
# We could avoid this line since we set the accelerator with `device_placement=True`.
batch.to(accelerator.device)
# New code #
# We use the new `accumulate` context manager to perform gradient accumulation
# We also currently do not support TPUs nor advise it as bugs were found on the XLA side when running our tests.
with accelerator.accumulate(__lowercase):
UpperCamelCase_ = model(**__lowercase)
UpperCamelCase_ = output.loss
accelerator.backward(__lowercase)
optimizer.step()
lr_scheduler.step()
optimizer.zero_grad()
model.eval()
for step, batch in enumerate(__lowercase):
# We could avoid this line since we set the accelerator with `device_placement=True`.
batch.to(accelerator.device)
with torch.no_grad():
UpperCamelCase_ = model(**__lowercase)
UpperCamelCase_ = outputs.logits.argmax(dim=-1)
UpperCamelCase_ , UpperCamelCase_ = accelerator.gather_for_metrics((predictions, batch['labels']))
metric.add_batch(
predictions=__lowercase , references=__lowercase , )
UpperCamelCase_ = metric.compute()
# Use accelerator.print to print only on the main process.
accelerator.print(f"""epoch {epoch}:""" , __lowercase)
def _snake_case ():
UpperCamelCase_ = argparse.ArgumentParser(description='Simple example of training script.')
parser.add_argument(
'--mixed_precision' , type=__lowercase , default=__lowercase , choices=['no', 'fp16', 'bf16', 'fp8'] , help='Whether to use mixed precision. Choose'
'between fp16 and bf16 (bfloat16). Bf16 requires PyTorch >= 1.10.'
'and an Nvidia Ampere GPU.' , )
# New Code #
parser.add_argument(
'--gradient_accumulation_steps' , type=__lowercase , default=1 , help='The number of minibatches to be ran before gradients are accumulated.' , )
parser.add_argument('--cpu' , action='store_true' , help='If passed, will train on the CPU.')
UpperCamelCase_ = parser.parse_args()
UpperCamelCase_ = {'lr': 2e-5, 'num_epochs': 3, 'seed': 42, 'batch_size': 16}
training_function(__lowercase , __lowercase)
if __name__ == "__main__":
main()
| 23 |
from typing import TYPE_CHECKING
from ...utils import OptionalDependencyNotAvailable, _LazyModule, is_torch_available
snake_case__ : Optional[int] = {
"""configuration_pegasus_x""": ["""PEGASUS_X_PRETRAINED_CONFIG_ARCHIVE_MAP""", """PegasusXConfig"""],
}
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
snake_case__ : Dict = [
"""PEGASUS_X_PRETRAINED_MODEL_ARCHIVE_LIST""",
"""PegasusXForConditionalGeneration""",
"""PegasusXModel""",
"""PegasusXPreTrainedModel""",
]
if TYPE_CHECKING:
from .configuration_pegasus_x import PEGASUS_X_PRETRAINED_CONFIG_ARCHIVE_MAP, PegasusXConfig
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_pegasus_x import (
PEGASUS_X_PRETRAINED_MODEL_ARCHIVE_LIST,
PegasusXForConditionalGeneration,
PegasusXModel,
PegasusXPreTrainedModel,
)
else:
import sys
snake_case__ : Optional[Any] = _LazyModule(__name__, globals()["""__file__"""], _import_structure, module_spec=__spec__)
| 23 | 1 |
# A Bipartite Graph is a graph whose vertices can be divided into two independent sets,
# U and V such that every edge (u, v) either connects a vertex from U to V or a vertex
# from V to U. In other words, for every edge (u, v), either u belongs to U and v to V,
# or u belongs to V and v to U. We can also say that there is no edge that connects
# vertices of same set.
def _snake_case (__lowercase):
UpperCamelCase_ = [False] * len(__lowercase)
UpperCamelCase_ = [-1] * len(__lowercase)
def dfs(__lowercase , __lowercase):
UpperCamelCase_ = True
UpperCamelCase_ = c
for u in graph[v]:
if not visited[u]:
dfs(__lowercase , 1 - c)
for i in range(len(__lowercase)):
if not visited[i]:
dfs(__lowercase , 0)
for i in range(len(__lowercase)):
for j in graph[i]:
if color[i] == color[j]:
return False
return True
# Adjacency list of graph
snake_case__ : Any = {0: [1, 3], 1: [0, 2], 2: [1, 3], 3: [0, 2], 4: []}
print(check_bipartite_dfs(graph))
| 23 |
import datasets
from .evaluate import evaluate
snake_case__ : int = """\
@article{hendrycks2021cuad,
title={CUAD: An Expert-Annotated NLP Dataset for Legal Contract Review},
author={Dan Hendrycks and Collin Burns and Anya Chen and Spencer Ball},
journal={arXiv preprint arXiv:2103.06268},
year={2021}
}
"""
snake_case__ : Union[str, Any] = """
This metric wrap the official scoring script for version 1 of the Contract
Understanding Atticus Dataset (CUAD).
Contract Understanding Atticus Dataset (CUAD) v1 is a corpus of more than 13,000 labels in 510
commercial legal contracts that have been manually labeled to identify 41 categories of important
clauses that lawyers look for when reviewing contracts in connection with corporate transactions.
"""
snake_case__ : Any = """
Computes CUAD scores (EM, F1, AUPR, Precision@80%Recall, and Precision@90%Recall).
Args:
predictions: List of question-answers dictionaries with the following key-values:
- 'id': id of the question-answer pair as given in the references (see below)
- 'prediction_text': list of possible texts for the answer, as a list of strings
depending on a threshold on the confidence probability of each prediction.
references: List of question-answers dictionaries with the following key-values:
- 'id': id of the question-answer pair (see above),
- 'answers': a Dict in the CUAD dataset format
{
'text': list of possible texts for the answer, as a list of strings
'answer_start': list of start positions for the answer, as a list of ints
}
Note that answer_start values are not taken into account to compute the metric.
Returns:
'exact_match': Exact match (the normalized answer exactly match the gold answer)
'f1': The F-score of predicted tokens versus the gold answer
'aupr': Area Under the Precision-Recall curve
'prec_at_80_recall': Precision at 80% recall
'prec_at_90_recall': Precision at 90% recall
Examples:
>>> predictions = [{'prediction_text': ['The seller:', 'The buyer/End-User: Shenzhen LOHAS Supply Chain Management Co., Ltd.'], 'id': 'LohaCompanyltd_20191209_F-1_EX-10.16_11917878_EX-10.16_Supply Agreement__Parties'}]
>>> references = [{'answers': {'answer_start': [143, 49], 'text': ['The seller:', 'The buyer/End-User: Shenzhen LOHAS Supply Chain Management Co., Ltd.']}, 'id': 'LohaCompanyltd_20191209_F-1_EX-10.16_11917878_EX-10.16_Supply Agreement__Parties'}]
>>> cuad_metric = datasets.load_metric(\"cuad\")
>>> results = cuad_metric.compute(predictions=predictions, references=references)
>>> print(results)
{'exact_match': 100.0, 'f1': 100.0, 'aupr': 0.0, 'prec_at_80_recall': 1.0, 'prec_at_90_recall': 1.0}
"""
@datasets.utils.file_utils.add_start_docstrings(_DESCRIPTION , _KWARGS_DESCRIPTION )
class _a ( datasets.Metric ):
"""simple docstring"""
def _UpperCAmelCase ( self ) -> Optional[int]:
return datasets.MetricInfo(
description=_DESCRIPTION , citation=_CITATION , inputs_description=_KWARGS_DESCRIPTION , features=datasets.Features(
{
'predictions': {
'id': datasets.Value('string' ),
'prediction_text': datasets.features.Sequence(datasets.Value('string' ) ),
},
'references': {
'id': datasets.Value('string' ),
'answers': datasets.features.Sequence(
{
'text': datasets.Value('string' ),
'answer_start': datasets.Value('int32' ),
} ),
},
} ) , codebase_urls=['https://www.atticusprojectai.org/cuad'] , reference_urls=['https://www.atticusprojectai.org/cuad'] , )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> Dict:
UpperCamelCase_ = {prediction['id']: prediction['prediction_text'] for prediction in predictions}
UpperCamelCase_ = [
{
'paragraphs': [
{
'qas': [
{
'answers': [{'text': answer_text} for answer_text in ref['answers']['text']],
'id': ref['id'],
}
for ref in references
]
}
]
}
]
UpperCamelCase_ = evaluate(dataset=_UpperCAmelCase , predictions=_UpperCAmelCase )
return score
| 23 | 1 |
import argparse
import re
import requests
import torch
# git clone https://github.com/salesforce/BLIP.git
from models.blip import blip_decoder
from models.blip_itm import blip_itm
from models.blip_vqa import blip_vqa
from PIL import Image
from torchvision import transforms
from torchvision.transforms.functional import InterpolationMode
from transformers import (
BertTokenizer,
BlipConfig,
BlipForConditionalGeneration,
BlipForImageTextRetrieval,
BlipForQuestionAnswering,
)
def _snake_case (__lowercase , __lowercase):
UpperCamelCase_ = 'https://storage.googleapis.com/sfr-vision-language-research/BLIP/demo.jpg'
UpperCamelCase_ = Image.open(requests.get(__lowercase , stream=__lowercase).raw).convert('RGB')
UpperCamelCase_ = transforms.Compose(
[
transforms.Resize((image_size, image_size) , interpolation=InterpolationMode.BICUBIC),
transforms.ToTensor(),
transforms.Normalize((0.48_145_466, 0.4_578_275, 0.40_821_073) , (0.26_862_954, 0.26_130_258, 0.27_577_711)),
])
UpperCamelCase_ = transform(__lowercase).unsqueeze(0).to(__lowercase)
return image
def _snake_case (__lowercase):
if "visual_encoder" in key:
UpperCamelCase_ = re.sub('visual_encoder*' , 'vision_model.encoder' , __lowercase)
if "blocks" in key:
UpperCamelCase_ = re.sub(r'blocks' , 'layers' , __lowercase)
if "attn" in key:
UpperCamelCase_ = re.sub(r'attn' , 'self_attn' , __lowercase)
if "norm1" in key:
UpperCamelCase_ = re.sub(r'norm1' , 'layer_norm1' , __lowercase)
if "norm2" in key:
UpperCamelCase_ = re.sub(r'norm2' , 'layer_norm2' , __lowercase)
if "encoder.norm" in key:
UpperCamelCase_ = re.sub(r'encoder.norm' , 'post_layernorm' , __lowercase)
if "encoder.patch_embed.proj" in key:
UpperCamelCase_ = re.sub(r'encoder.patch_embed.proj' , 'embeddings.patch_embedding' , __lowercase)
if "encoder.pos_embed" in key:
UpperCamelCase_ = re.sub(r'encoder.pos_embed' , 'embeddings.position_embedding' , __lowercase)
if "encoder.cls_token" in key:
UpperCamelCase_ = re.sub(r'encoder.cls_token' , 'embeddings.class_embedding' , __lowercase)
if "self_attn" in key:
UpperCamelCase_ = re.sub(r'self_attn.proj' , 'self_attn.projection' , __lowercase)
return key
@torch.no_grad()
def _snake_case (__lowercase , __lowercase=None):
if config_path is not None:
UpperCamelCase_ = BlipConfig.from_pretrained(__lowercase)
else:
UpperCamelCase_ = BlipConfig(projection_dim=512 , text_config={} , vision_config={})
UpperCamelCase_ = BlipForConditionalGeneration(__lowercase).eval()
UpperCamelCase_ = 'https://storage.googleapis.com/sfr-vision-language-research/BLIP/models/model_base_capfilt_large.pth'
UpperCamelCase_ = blip_decoder(pretrained=__lowercase , image_size=384 , vit='base')
UpperCamelCase_ = pt_model.eval()
UpperCamelCase_ = pt_model.state_dict()
for key in modified_state_dict.copy():
UpperCamelCase_ = modified_state_dict.pop(__lowercase)
UpperCamelCase_ = rename_key(__lowercase)
UpperCamelCase_ = value
hf_model.load_state_dict(__lowercase)
UpperCamelCase_ = 384
UpperCamelCase_ = load_demo_image(image_size=__lowercase , device='cpu')
UpperCamelCase_ = BertTokenizer.from_pretrained('bert-base-uncased')
UpperCamelCase_ = tokenizer(['a picture of']).input_ids
UpperCamelCase_ = hf_model.generate(__lowercase , __lowercase)
assert out[0].tolist() == [30522, 1037, 3861, 1997, 1037, 2450, 3564, 2006, 1996, 3509, 2007, 2014, 3899, 102]
UpperCamelCase_ = hf_model.generate(__lowercase)
assert out[0].tolist() == [30522, 1037, 2450, 3564, 2006, 1996, 3509, 2007, 2014, 3899, 102]
if pytorch_dump_folder_path is not None:
hf_model.save_pretrained(__lowercase)
# model_url = 'https://storage.googleapis.com/sfr-vision-language-research/BLIP/models/model_vqa.pth'
UpperCamelCase_ = (
'https://storage.googleapis.com/sfr-vision-language-research/BLIP/models/model_base_vqa_capfilt_large.pth'
)
UpperCamelCase_ = blip_vqa(pretrained=__lowercase , image_size=__lowercase , vit='base')
vqa_model.eval()
UpperCamelCase_ = vqa_model.state_dict()
for key in modified_state_dict.copy():
UpperCamelCase_ = modified_state_dict.pop(__lowercase)
UpperCamelCase_ = rename_key(__lowercase)
UpperCamelCase_ = value
UpperCamelCase_ = BlipForQuestionAnswering(__lowercase)
hf_vqa_model.load_state_dict(__lowercase)
UpperCamelCase_ = ['How many dogs are in this image?']
UpperCamelCase_ = tokenizer(__lowercase , return_tensors='pt').input_ids
UpperCamelCase_ = hf_vqa_model.generate(__lowercase , __lowercase)
print(tokenizer.decode(answer[0]))
assert tokenizer.decode(answer[0]) == "[UNK] 1 [SEP]"
if pytorch_dump_folder_path is not None:
hf_vqa_model.save_pretrained(pytorch_dump_folder_path + '_vqa')
UpperCamelCase_ = 'https://storage.googleapis.com/sfr-vision-language-research/BLIP/models/model_base_retrieval_coco.pth'
UpperCamelCase_ = blip_itm(pretrained=__lowercase , image_size=__lowercase , vit='base')
itm_model.eval()
UpperCamelCase_ = itm_model.state_dict()
for key in modified_state_dict.copy():
UpperCamelCase_ = modified_state_dict.pop(__lowercase)
UpperCamelCase_ = rename_key(__lowercase)
UpperCamelCase_ = value
UpperCamelCase_ = BlipForImageTextRetrieval(__lowercase)
UpperCamelCase_ = ['A picture of a woman with a dog sitting in a beach']
UpperCamelCase_ = tokenizer(
__lowercase , return_tensors='pt' , padding='max_length' , truncation=__lowercase , max_length=35 , ).input_ids
hf_itm_model.load_state_dict(__lowercase)
hf_itm_model.eval()
UpperCamelCase_ = hf_itm_model(__lowercase , __lowercase , use_itm_head=__lowercase)
UpperCamelCase_ = hf_itm_model(__lowercase , __lowercase , use_itm_head=__lowercase)
assert out[0].item() == 0.2_110_687_494_277_954
assert torch.nn.functional.softmax(out_itm[0] , dim=1)[:, 1].item() == 0.45_698_845_386_505_127
if pytorch_dump_folder_path is not None:
hf_itm_model.save_pretrained(pytorch_dump_folder_path + '_itm')
if __name__ == "__main__":
snake_case__ : Tuple = argparse.ArgumentParser()
parser.add_argument("""--pytorch_dump_folder_path""", default=None, type=str, help="""Path to the output PyTorch model.""")
parser.add_argument("""--config_path""", default=None, type=str, help="""Path to hf config.json of model to convert""")
snake_case__ : Optional[Any] = parser.parse_args()
convert_blip_checkpoint(args.checkpoint_path, args.pytorch_dump_folder_path, args.config_path)
| 23 |
import os
import tempfile
from functools import partial
from unittest import TestCase
from unittest.mock import patch
import datasets
import datasets.config
from .utils import require_beam
class _a ( datasets.BeamBasedBuilder ):
"""simple docstring"""
def _UpperCAmelCase ( self ) -> List[str]:
return datasets.DatasetInfo(
features=datasets.Features({'content': datasets.Value('string' )} ) , supervised_keys=_UpperCAmelCase , )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> Optional[Any]:
return [datasets.SplitGenerator(name=datasets.Split.TRAIN , gen_kwargs={'examples': get_test_dummy_examples()} )]
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> Tuple:
import apache_beam as beam
return pipeline | "Load Examples" >> beam.Create(_UpperCAmelCase )
class _a ( datasets.BeamBasedBuilder ):
"""simple docstring"""
def _UpperCAmelCase ( self ) -> Any:
return datasets.DatasetInfo(
features=datasets.Features({'a': datasets.Sequence({'b': datasets.Value('string' )} )} ) , supervised_keys=_UpperCAmelCase , )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> Tuple:
return [
datasets.SplitGenerator(name=datasets.Split.TRAIN , gen_kwargs={'examples': get_test_nested_examples()} )
]
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> Union[str, Any]:
import apache_beam as beam
return pipeline | "Load Examples" >> beam.Create(_UpperCAmelCase )
def _snake_case ():
return [(i, {"content": content}) for i, content in enumerate(['foo', 'bar', 'foobar'])]
def _snake_case ():
return [(i, {"a": {"b": [content]}}) for i, content in enumerate(['foo', 'bar', 'foobar'])]
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
@require_beam
def _UpperCAmelCase ( self ) -> Dict:
UpperCamelCase_ = len(get_test_dummy_examples() )
with tempfile.TemporaryDirectory() as tmp_cache_dir:
UpperCamelCase_ = DummyBeamDataset(cache_dir=_UpperCAmelCase , beam_runner='DirectRunner' )
builder.download_and_prepare()
self.assertTrue(
os.path.exists(
os.path.join(_UpperCAmelCase , builder.name , 'default' , '0.0.0' , f"""{builder.name}-train.arrow""" ) ) )
self.assertDictEqual(builder.info.features , datasets.Features({'content': datasets.Value('string' )} ) )
UpperCamelCase_ = builder.as_dataset()
self.assertEqual(dset['train'].num_rows , _UpperCAmelCase )
self.assertEqual(dset['train'].info.splits['train'].num_examples , _UpperCAmelCase )
self.assertDictEqual(dset['train'][0] , get_test_dummy_examples()[0][1] )
self.assertDictEqual(
dset['train'][expected_num_examples - 1] , get_test_dummy_examples()[expected_num_examples - 1][1] )
self.assertTrue(
os.path.exists(os.path.join(_UpperCAmelCase , builder.name , 'default' , '0.0.0' , 'dataset_info.json' ) ) )
del dset
@require_beam
def _UpperCAmelCase ( self ) -> List[str]:
import apache_beam as beam
UpperCamelCase_ = beam.io.parquetio.WriteToParquet
UpperCamelCase_ = len(get_test_dummy_examples() )
with tempfile.TemporaryDirectory() as tmp_cache_dir:
UpperCamelCase_ = DummyBeamDataset(cache_dir=_UpperCAmelCase , beam_runner='DirectRunner' )
with patch('apache_beam.io.parquetio.WriteToParquet' ) as write_parquet_mock:
UpperCamelCase_ = partial(_UpperCAmelCase , num_shards=2 )
builder.download_and_prepare()
self.assertTrue(
os.path.exists(
os.path.join(
_UpperCAmelCase , builder.name , 'default' , '0.0.0' , f"""{builder.name}-train-00000-of-00002.arrow""" ) ) )
self.assertTrue(
os.path.exists(
os.path.join(
_UpperCAmelCase , builder.name , 'default' , '0.0.0' , f"""{builder.name}-train-00000-of-00002.arrow""" ) ) )
self.assertDictEqual(builder.info.features , datasets.Features({'content': datasets.Value('string' )} ) )
UpperCamelCase_ = builder.as_dataset()
self.assertEqual(dset['train'].num_rows , _UpperCAmelCase )
self.assertEqual(dset['train'].info.splits['train'].num_examples , _UpperCAmelCase )
# Order is not preserved when sharding, so we just check that all the elements are there
self.assertListEqual(sorted(dset['train']['content'] ) , sorted(['foo', 'bar', 'foobar'] ) )
self.assertTrue(
os.path.exists(os.path.join(_UpperCAmelCase , builder.name , 'default' , '0.0.0' , 'dataset_info.json' ) ) )
del dset
@require_beam
def _UpperCAmelCase ( self ) -> Any:
with tempfile.TemporaryDirectory() as tmp_cache_dir:
UpperCamelCase_ = DummyBeamDataset(cache_dir=_UpperCAmelCase )
self.assertRaises(datasets.builder.MissingBeamOptions , builder.download_and_prepare )
@require_beam
def _UpperCAmelCase ( self ) -> Union[str, Any]:
UpperCamelCase_ = len(get_test_nested_examples() )
with tempfile.TemporaryDirectory() as tmp_cache_dir:
UpperCamelCase_ = NestedBeamDataset(cache_dir=_UpperCAmelCase , beam_runner='DirectRunner' )
builder.download_and_prepare()
self.assertTrue(
os.path.exists(
os.path.join(_UpperCAmelCase , builder.name , 'default' , '0.0.0' , f"""{builder.name}-train.arrow""" ) ) )
self.assertDictEqual(
builder.info.features , datasets.Features({'a': datasets.Sequence({'b': datasets.Value('string' )} )} ) )
UpperCamelCase_ = builder.as_dataset()
self.assertEqual(dset['train'].num_rows , _UpperCAmelCase )
self.assertEqual(dset['train'].info.splits['train'].num_examples , _UpperCAmelCase )
self.assertDictEqual(dset['train'][0] , get_test_nested_examples()[0][1] )
self.assertDictEqual(
dset['train'][expected_num_examples - 1] , get_test_nested_examples()[expected_num_examples - 1][1] )
self.assertTrue(
os.path.exists(os.path.join(_UpperCAmelCase , builder.name , 'default' , '0.0.0' , 'dataset_info.json' ) ) )
del dset
| 23 | 1 |
import os
from typing import Dict, List, Tuple, TypeVar, Union
snake_case__ : int = TypeVar("""T""")
snake_case__ : str = Union[List[T], Tuple[T, ...]]
snake_case__ : Any = Union[T, List[T], Dict[str, T]]
snake_case__ : Tuple = Union[str, bytes, os.PathLike]
| 23 |
import argparse
import torch
from ...utils import logging
from . import AlbertConfig, AlbertForPreTraining, load_tf_weights_in_albert
logging.set_verbosity_info()
def _snake_case (__lowercase , __lowercase , __lowercase):
# Initialise PyTorch model
UpperCamelCase_ = AlbertConfig.from_json_file(__lowercase)
print(f"""Building PyTorch model from configuration: {config}""")
UpperCamelCase_ = AlbertForPreTraining(__lowercase)
# Load weights from tf checkpoint
load_tf_weights_in_albert(__lowercase , __lowercase , __lowercase)
# Save pytorch-model
print(f"""Save PyTorch model to {pytorch_dump_path}""")
torch.save(model.state_dict() , __lowercase)
if __name__ == "__main__":
snake_case__ : Tuple = argparse.ArgumentParser()
# Required parameters
parser.add_argument(
"""--tf_checkpoint_path""", default=None, type=str, required=True, help="""Path to the TensorFlow checkpoint path."""
)
parser.add_argument(
"""--albert_config_file""",
default=None,
type=str,
required=True,
help=(
"""The config json file corresponding to the pre-trained ALBERT model. \n"""
"""This specifies the model architecture."""
),
)
parser.add_argument(
"""--pytorch_dump_path""", default=None, type=str, required=True, help="""Path to the output PyTorch model."""
)
snake_case__ : str = parser.parse_args()
convert_tf_checkpoint_to_pytorch(args.tf_checkpoint_path, args.albert_config_file, args.pytorch_dump_path)
| 23 | 1 |
from pathlib import Path
import cva
import numpy as np
from matplotlib import pyplot as plt
def _snake_case (__lowercase , __lowercase , __lowercase , __lowercase , __lowercase):
UpperCamelCase_ = cva.getAffineTransform(__lowercase , __lowercase)
return cva.warpAffine(__lowercase , __lowercase , (rows, cols))
if __name__ == "__main__":
# read original image
snake_case__ : Optional[int] = cva.imread(
str(Path(__file__).resolve().parent.parent / """image_data""" / """lena.jpg""")
)
# turn image in gray scale value
snake_case__ : Any = cva.cvtColor(image, cva.COLOR_BGR2GRAY)
# get image shape
snake_case__ , snake_case__ : Optional[Any] = gray_img.shape
# set different points to rotate image
snake_case__ : List[Any] = np.array([[5_0, 5_0], [2_0_0, 5_0], [5_0, 2_0_0]], np.floataa)
snake_case__ : Union[str, Any] = np.array([[1_0, 1_0_0], [2_0_0, 5_0], [1_0_0, 2_5_0]], np.floataa)
snake_case__ : Optional[int] = np.array([[5_0, 5_0], [1_5_0, 5_0], [1_2_0, 2_0_0]], np.floataa)
snake_case__ : int = np.array([[1_0, 1_0_0], [8_0, 5_0], [1_8_0, 2_5_0]], np.floataa)
# add all rotated images in a list
snake_case__ : Optional[Any] = [
gray_img,
get_rotation(gray_img, ptsa, ptsa, img_rows, img_cols),
get_rotation(gray_img, ptsa, ptsa, img_rows, img_cols),
get_rotation(gray_img, ptsa, ptsa, img_rows, img_cols),
]
# plot different image rotations
snake_case__ : Optional[int] = plt.figure(1)
snake_case__ : int = ["""Original""", """Rotation 1""", """Rotation 2""", """Rotation 3"""]
for i, image in enumerate(images):
plt.subplot(2, 2, i + 1), plt.imshow(image, """gray""")
plt.title(titles[i])
plt.axis("""off""")
plt.subplots_adjust(left=0.0, bottom=0.05, right=1.0, top=0.95)
plt.show()
| 23 |
from transformers import BertTokenizer, EncoderDecoderModel, SeqaSeqTrainer, SeqaSeqTrainingArguments
from transformers.testing_utils import TestCasePlus, require_torch, slow
from transformers.utils import is_datasets_available
if is_datasets_available():
import datasets
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
@slow
@require_torch
def _UpperCAmelCase ( self ) -> List[Any]:
UpperCamelCase_ = EncoderDecoderModel.from_encoder_decoder_pretrained('prajjwal1/bert-tiny' , 'prajjwal1/bert-tiny' )
UpperCamelCase_ = BertTokenizer.from_pretrained('bert-base-uncased' )
UpperCamelCase_ = bertabert.config.encoder.vocab_size
UpperCamelCase_ = tokenizer.sep_token_id
UpperCamelCase_ = tokenizer.cls_token_id
UpperCamelCase_ = 128
UpperCamelCase_ = datasets.load_dataset('cnn_dailymail' , '3.0.0' , split='train[:1%]' )
UpperCamelCase_ = datasets.load_dataset('cnn_dailymail' , '3.0.0' , split='validation[:1%]' )
UpperCamelCase_ = train_dataset.select(range(32 ) )
UpperCamelCase_ = val_dataset.select(range(16 ) )
UpperCamelCase_ = 4
def _map_to_encoder_decoder_inputs(_UpperCAmelCase ):
# Tokenizer will automatically set [BOS] <text> [EOS]
UpperCamelCase_ = tokenizer(batch['article'] , padding='max_length' , truncation=_UpperCAmelCase , max_length=512 )
UpperCamelCase_ = tokenizer(batch['highlights'] , padding='max_length' , truncation=_UpperCAmelCase , max_length=128 )
UpperCamelCase_ = inputs.input_ids
UpperCamelCase_ = inputs.attention_mask
UpperCamelCase_ = outputs.input_ids
UpperCamelCase_ = outputs.input_ids.copy()
UpperCamelCase_ = [
[-100 if token == tokenizer.pad_token_id else token for token in labels] for labels in batch['labels']
]
UpperCamelCase_ = outputs.attention_mask
assert all(len(_UpperCAmelCase ) == 512 for x in inputs.input_ids )
assert all(len(_UpperCAmelCase ) == 128 for x in outputs.input_ids )
return batch
def _compute_metrics(_UpperCAmelCase ):
UpperCamelCase_ = pred.label_ids
UpperCamelCase_ = pred.predictions
# all unnecessary tokens are removed
UpperCamelCase_ = tokenizer.batch_decode(_UpperCAmelCase , skip_special_tokens=_UpperCAmelCase )
UpperCamelCase_ = tokenizer.batch_decode(_UpperCAmelCase , skip_special_tokens=_UpperCAmelCase )
UpperCamelCase_ = sum([int(pred_str[i] == label_str[i] ) for i in range(len(_UpperCAmelCase ) )] ) / len(_UpperCAmelCase )
return {"accuracy": accuracy}
# map train dataset
UpperCamelCase_ = train_dataset.map(
_map_to_encoder_decoder_inputs , batched=_UpperCAmelCase , batch_size=_UpperCAmelCase , remove_columns=['article', 'highlights'] , )
train_dataset.set_format(
type='torch' , columns=['input_ids', 'attention_mask', 'decoder_input_ids', 'decoder_attention_mask', 'labels'] , )
# same for validation dataset
UpperCamelCase_ = val_dataset.map(
_map_to_encoder_decoder_inputs , batched=_UpperCAmelCase , batch_size=_UpperCAmelCase , remove_columns=['article', 'highlights'] , )
val_dataset.set_format(
type='torch' , columns=['input_ids', 'attention_mask', 'decoder_input_ids', 'decoder_attention_mask', 'labels'] , )
UpperCamelCase_ = self.get_auto_remove_tmp_dir()
UpperCamelCase_ = SeqaSeqTrainingArguments(
output_dir=_UpperCAmelCase , per_device_train_batch_size=_UpperCAmelCase , per_device_eval_batch_size=_UpperCAmelCase , predict_with_generate=_UpperCAmelCase , evaluation_strategy='steps' , do_train=_UpperCAmelCase , do_eval=_UpperCAmelCase , warmup_steps=0 , eval_steps=2 , logging_steps=2 , )
# instantiate trainer
UpperCamelCase_ = SeqaSeqTrainer(
model=_UpperCAmelCase , args=_UpperCAmelCase , compute_metrics=_compute_metrics , train_dataset=_UpperCAmelCase , eval_dataset=_UpperCAmelCase , tokenizer=_UpperCAmelCase , )
# start training
trainer.train()
| 23 | 1 |
from typing import List, Union
from ..utils import (
add_end_docstrings,
is_tf_available,
is_torch_available,
is_vision_available,
logging,
requires_backends,
)
from .base import PIPELINE_INIT_ARGS, Pipeline
if is_vision_available():
from PIL import Image
from ..image_utils import load_image
if is_tf_available():
from ..models.auto.modeling_tf_auto import TF_MODEL_FOR_VISION_2_SEQ_MAPPING
if is_torch_available():
import torch
from ..models.auto.modeling_auto import MODEL_FOR_VISION_2_SEQ_MAPPING
snake_case__ : Union[str, Any] = logging.get_logger(__name__)
@add_end_docstrings(UpperCAmelCase__ )
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
def __init__( self , *_UpperCAmelCase , **_UpperCAmelCase ) -> Any:
super().__init__(*_UpperCAmelCase , **_UpperCAmelCase )
requires_backends(self , 'vision' )
self.check_model_type(
TF_MODEL_FOR_VISION_2_SEQ_MAPPING if self.framework == 'tf' else MODEL_FOR_VISION_2_SEQ_MAPPING )
def _UpperCAmelCase ( self , _UpperCAmelCase=None , _UpperCAmelCase=None , _UpperCAmelCase=None ) -> List[str]:
UpperCamelCase_ = {}
UpperCamelCase_ = {}
if prompt is not None:
UpperCamelCase_ = prompt
if generate_kwargs is not None:
UpperCamelCase_ = generate_kwargs
if max_new_tokens is not None:
if "generate_kwargs" not in forward_kwargs:
UpperCamelCase_ = {}
if "max_new_tokens" in forward_kwargs["generate_kwargs"]:
raise ValueError(
'\'max_new_tokens\' is defined twice, once in \'generate_kwargs\' and once as a direct parameter,'
' please use only one' )
UpperCamelCase_ = max_new_tokens
return preprocess_params, forward_kwargs, {}
def __call__( self , _UpperCAmelCase , **_UpperCAmelCase ) -> int:
return super().__call__(_UpperCAmelCase , **_UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase=None ) -> str:
UpperCamelCase_ = load_image(_UpperCAmelCase )
if prompt is not None:
if not isinstance(_UpperCAmelCase , _UpperCAmelCase ):
raise ValueError(
f"""Received an invalid text input, got - {type(_UpperCAmelCase )} - but expected a single string. """
'Note also that one single text can be provided for conditional image to text generation.' )
UpperCamelCase_ = self.model.config.model_type
if model_type == "git":
UpperCamelCase_ = self.image_processor(images=_UpperCAmelCase , return_tensors=self.framework )
UpperCamelCase_ = self.tokenizer(text=_UpperCAmelCase , add_special_tokens=_UpperCAmelCase ).input_ids
UpperCamelCase_ = [self.tokenizer.cls_token_id] + input_ids
UpperCamelCase_ = torch.tensor(_UpperCAmelCase ).unsqueeze(0 )
model_inputs.update({'input_ids': input_ids} )
elif model_type == "pix2struct":
UpperCamelCase_ = self.image_processor(images=_UpperCAmelCase , header_text=_UpperCAmelCase , return_tensors=self.framework )
elif model_type != "vision-encoder-decoder":
# vision-encoder-decoder does not support conditional generation
UpperCamelCase_ = self.image_processor(images=_UpperCAmelCase , return_tensors=self.framework )
UpperCamelCase_ = self.tokenizer(_UpperCAmelCase , return_tensors=self.framework )
model_inputs.update(_UpperCAmelCase )
else:
raise ValueError(f"""Model type {model_type} does not support conditional text generation""" )
else:
UpperCamelCase_ = self.image_processor(images=_UpperCAmelCase , return_tensors=self.framework )
if self.model.config.model_type == "git" and prompt is None:
UpperCamelCase_ = None
return model_inputs
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase=None ) -> Any:
# Git model sets `model_inputs["input_ids"] = None` in `preprocess` (when `prompt=None`). In batch model, the
# pipeline will group them into a list of `None`, which fail `_forward`. Avoid this by checking it first.
if (
"input_ids" in model_inputs
and isinstance(model_inputs['input_ids'] , _UpperCAmelCase )
and all(x is None for x in model_inputs['input_ids'] )
):
UpperCamelCase_ = None
if generate_kwargs is None:
UpperCamelCase_ = {}
# FIXME: We need to pop here due to a difference in how `generation.py` and `generation.tf_utils.py`
# parse inputs. In the Tensorflow version, `generate` raises an error if we don't use `input_ids` whereas
# the PyTorch version matches it with `self.model.main_input_name` or `self.model.encoder.main_input_name`
# in the `_prepare_model_inputs` method.
UpperCamelCase_ = model_inputs.pop(self.model.main_input_name )
UpperCamelCase_ = self.model.generate(_UpperCAmelCase , **_UpperCAmelCase , **_UpperCAmelCase )
return model_outputs
def _UpperCAmelCase ( self , _UpperCAmelCase ) -> Any:
UpperCamelCase_ = []
for output_ids in model_outputs:
UpperCamelCase_ = {
'generated_text': self.tokenizer.decode(
_UpperCAmelCase , skip_special_tokens=_UpperCAmelCase , )
}
records.append(_UpperCAmelCase )
return records
| 23 |
import argparse
import os
import evaluate
import torch
from datasets import load_dataset
from torch.optim import AdamW
from torch.utils.data import DataLoader
from transformers import AutoModelForSequenceClassification, AutoTokenizer, get_linear_schedule_with_warmup, set_seed
from accelerate import Accelerator, DistributedType
########################################################################
# This is a fully working simple example to use Accelerate
# and perform gradient accumulation
#
# This example trains a Bert base model on GLUE MRPC
# in any of the following settings (with the same script):
# - single CPU or single GPU
# - multi GPUS (using PyTorch distributed mode)
# - (multi) TPUs
# - fp16 (mixed-precision) or fp32 (normal precision)
#
# To run it in each of these various modes, follow the instructions
# in the readme for examples:
# https://github.com/huggingface/accelerate/tree/main/examples
#
########################################################################
snake_case__ : Dict = 1_6
snake_case__ : List[str] = 3_2
def _snake_case (__lowercase , __lowercase = 16):
UpperCamelCase_ = AutoTokenizer.from_pretrained('bert-base-cased')
UpperCamelCase_ = load_dataset('glue' , 'mrpc')
def tokenize_function(__lowercase):
# max_length=None => use the model max length (it's actually the default)
UpperCamelCase_ = tokenizer(examples['sentence1'] , examples['sentence2'] , truncation=__lowercase , max_length=__lowercase)
return outputs
# Apply the method we just defined to all the examples in all the splits of the dataset
# starting with the main process first:
with accelerator.main_process_first():
UpperCamelCase_ = datasets.map(
__lowercase , batched=__lowercase , remove_columns=['idx', 'sentence1', 'sentence2'] , )
# We also rename the 'label' column to 'labels' which is the expected name for labels by the models of the
# transformers library
UpperCamelCase_ = tokenized_datasets.rename_column('label' , 'labels')
def collate_fn(__lowercase):
# On TPU it's best to pad everything to the same length or training will be very slow.
UpperCamelCase_ = 128 if accelerator.distributed_type == DistributedType.TPU else None
# When using mixed precision we want round multiples of 8/16
if accelerator.mixed_precision == "fp8":
UpperCamelCase_ = 16
elif accelerator.mixed_precision != "no":
UpperCamelCase_ = 8
else:
UpperCamelCase_ = None
return tokenizer.pad(
__lowercase , padding='longest' , max_length=__lowercase , pad_to_multiple_of=__lowercase , return_tensors='pt' , )
# Instantiate dataloaders.
UpperCamelCase_ = DataLoader(
tokenized_datasets['train'] , shuffle=__lowercase , collate_fn=__lowercase , batch_size=__lowercase)
UpperCamelCase_ = DataLoader(
tokenized_datasets['validation'] , shuffle=__lowercase , collate_fn=__lowercase , batch_size=__lowercase)
return train_dataloader, eval_dataloader
# For testing only
if os.environ.get("""TESTING_MOCKED_DATALOADERS""", None) == "1":
from accelerate.test_utils.training import mocked_dataloaders
snake_case__ : List[str] = mocked_dataloaders # noqa: F811
def _snake_case (__lowercase , __lowercase):
# For testing only
if os.environ.get('TESTING_MOCKED_DATALOADERS' , __lowercase) == "1":
UpperCamelCase_ = 2
# New Code #
UpperCamelCase_ = int(args.gradient_accumulation_steps)
# Initialize accelerator
UpperCamelCase_ = Accelerator(
cpu=args.cpu , mixed_precision=args.mixed_precision , gradient_accumulation_steps=__lowercase)
if accelerator.distributed_type == DistributedType.TPU and gradient_accumulation_steps > 1:
raise NotImplementedError(
'Gradient accumulation on TPUs is currently not supported. Pass `gradient_accumulation_steps=1`')
# Sample hyper-parameters for learning rate, batch size, seed and a few other HPs
UpperCamelCase_ = config['lr']
UpperCamelCase_ = int(config['num_epochs'])
UpperCamelCase_ = int(config['seed'])
UpperCamelCase_ = int(config['batch_size'])
UpperCamelCase_ = evaluate.load('glue' , 'mrpc')
set_seed(__lowercase)
UpperCamelCase_ , UpperCamelCase_ = get_dataloaders(__lowercase , __lowercase)
# Instantiate the model (we build the model here so that the seed also control new weights initialization)
UpperCamelCase_ = AutoModelForSequenceClassification.from_pretrained('bert-base-cased' , return_dict=__lowercase)
# We could avoid this line since the accelerator is set with `device_placement=True` (default value).
# Note that if you are placing tensors on devices manually, this line absolutely needs to be before the optimizer
# creation otherwise training will not work on TPU (`accelerate` will kindly throw an error to make us aware of that).
UpperCamelCase_ = model.to(accelerator.device)
# Instantiate optimizer
UpperCamelCase_ = AdamW(params=model.parameters() , lr=__lowercase)
# Instantiate scheduler
UpperCamelCase_ = get_linear_schedule_with_warmup(
optimizer=__lowercase , num_warmup_steps=100 , num_training_steps=(len(__lowercase) * num_epochs) , )
# Prepare everything
# There is no specific order to remember, we just need to unpack the objects in the same order we gave them to the
# prepare method.
UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ = accelerator.prepare(
__lowercase , __lowercase , __lowercase , __lowercase , __lowercase)
# Now we train the model
for epoch in range(__lowercase):
model.train()
for step, batch in enumerate(__lowercase):
# We could avoid this line since we set the accelerator with `device_placement=True`.
batch.to(accelerator.device)
# New code #
# We use the new `accumulate` context manager to perform gradient accumulation
# We also currently do not support TPUs nor advise it as bugs were found on the XLA side when running our tests.
with accelerator.accumulate(__lowercase):
UpperCamelCase_ = model(**__lowercase)
UpperCamelCase_ = output.loss
accelerator.backward(__lowercase)
optimizer.step()
lr_scheduler.step()
optimizer.zero_grad()
model.eval()
for step, batch in enumerate(__lowercase):
# We could avoid this line since we set the accelerator with `device_placement=True`.
batch.to(accelerator.device)
with torch.no_grad():
UpperCamelCase_ = model(**__lowercase)
UpperCamelCase_ = outputs.logits.argmax(dim=-1)
UpperCamelCase_ , UpperCamelCase_ = accelerator.gather_for_metrics((predictions, batch['labels']))
metric.add_batch(
predictions=__lowercase , references=__lowercase , )
UpperCamelCase_ = metric.compute()
# Use accelerator.print to print only on the main process.
accelerator.print(f"""epoch {epoch}:""" , __lowercase)
def _snake_case ():
UpperCamelCase_ = argparse.ArgumentParser(description='Simple example of training script.')
parser.add_argument(
'--mixed_precision' , type=__lowercase , default=__lowercase , choices=['no', 'fp16', 'bf16', 'fp8'] , help='Whether to use mixed precision. Choose'
'between fp16 and bf16 (bfloat16). Bf16 requires PyTorch >= 1.10.'
'and an Nvidia Ampere GPU.' , )
# New Code #
parser.add_argument(
'--gradient_accumulation_steps' , type=__lowercase , default=1 , help='The number of minibatches to be ran before gradients are accumulated.' , )
parser.add_argument('--cpu' , action='store_true' , help='If passed, will train on the CPU.')
UpperCamelCase_ = parser.parse_args()
UpperCamelCase_ = {'lr': 2e-5, 'num_epochs': 3, 'seed': 42, 'batch_size': 16}
training_function(__lowercase , __lowercase)
if __name__ == "__main__":
main()
| 23 | 1 |
import os
import sys
from contextlib import contextmanager
# Windows only
if os.name == "nt":
import ctypes
import msvcrt # noqa
class _a ( ctypes.Structure ):
"""simple docstring"""
A_ = [("""size""", ctypes.c_int), ("""visible""", ctypes.c_byte)]
def _snake_case ():
if os.name == "nt":
UpperCamelCase_ = CursorInfo()
UpperCamelCase_ = ctypes.windll.kernelaa.GetStdHandle(-11)
ctypes.windll.kernelaa.GetConsoleCursorInfo(__lowercase , ctypes.byref(__lowercase))
UpperCamelCase_ = False
ctypes.windll.kernelaa.SetConsoleCursorInfo(__lowercase , ctypes.byref(__lowercase))
elif os.name == "posix":
sys.stdout.write('\033[?25l')
sys.stdout.flush()
def _snake_case ():
if os.name == "nt":
UpperCamelCase_ = CursorInfo()
UpperCamelCase_ = ctypes.windll.kernelaa.GetStdHandle(-11)
ctypes.windll.kernelaa.GetConsoleCursorInfo(__lowercase , ctypes.byref(__lowercase))
UpperCamelCase_ = True
ctypes.windll.kernelaa.SetConsoleCursorInfo(__lowercase , ctypes.byref(__lowercase))
elif os.name == "posix":
sys.stdout.write('\033[?25h')
sys.stdout.flush()
@contextmanager
def _snake_case ():
try:
hide_cursor()
yield
finally:
show_cursor()
| 23 |
import inspect
import unittest
import numpy as np
from tests.test_modeling_common import floats_tensor
from transformers import MaskaFormerConfig, is_torch_available, is_vision_available
from transformers.testing_utils import require_torch, require_torch_multi_gpu, require_vision, slow, torch_device
from transformers.utils import cached_property
from ...test_configuration_common import ConfigTester
from ...test_modeling_common import ModelTesterMixin
from ...test_pipeline_mixin import PipelineTesterMixin
if is_torch_available():
import torch
from transformers import MaskaFormerForUniversalSegmentation, MaskaFormerModel
if is_vision_available():
from transformers import MaskaFormerImageProcessor
if is_vision_available():
from PIL import Image
class _a :
"""simple docstring"""
def __init__( self , _UpperCAmelCase , _UpperCAmelCase=2 , _UpperCAmelCase=True , _UpperCAmelCase=False , _UpperCAmelCase=10 , _UpperCAmelCase=3 , _UpperCAmelCase=32 * 8 , _UpperCAmelCase=32 * 8 , _UpperCAmelCase=4 , _UpperCAmelCase=64 , ) -> List[Any]:
UpperCamelCase_ = parent
UpperCamelCase_ = batch_size
UpperCamelCase_ = is_training
UpperCamelCase_ = use_auxiliary_loss
UpperCamelCase_ = num_queries
UpperCamelCase_ = num_channels
UpperCamelCase_ = min_size
UpperCamelCase_ = max_size
UpperCamelCase_ = num_labels
UpperCamelCase_ = hidden_dim
UpperCamelCase_ = hidden_dim
def _UpperCAmelCase ( self ) -> List[str]:
UpperCamelCase_ = floats_tensor([self.batch_size, self.num_channels, self.min_size, self.max_size] ).to(
_UpperCAmelCase )
UpperCamelCase_ = torch.ones([self.batch_size, self.min_size, self.max_size] , device=_UpperCAmelCase )
UpperCamelCase_ = (
torch.rand([self.batch_size, self.num_labels, self.min_size, self.max_size] , device=_UpperCAmelCase ) > 0.5
).float()
UpperCamelCase_ = (torch.rand((self.batch_size, self.num_labels) , device=_UpperCAmelCase ) > 0.5).long()
UpperCamelCase_ = self.get_config()
return config, pixel_values, pixel_mask, mask_labels, class_labels
def _UpperCAmelCase ( self ) -> Optional[int]:
UpperCamelCase_ = MaskaFormerConfig(
hidden_size=self.hidden_dim , )
UpperCamelCase_ = self.num_queries
UpperCamelCase_ = self.num_labels
UpperCamelCase_ = [1, 1, 1, 1]
UpperCamelCase_ = self.num_channels
UpperCamelCase_ = 64
UpperCamelCase_ = 128
UpperCamelCase_ = self.hidden_dim
UpperCamelCase_ = self.hidden_dim
UpperCamelCase_ = self.hidden_dim
return config
def _UpperCAmelCase ( self ) -> Dict:
UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ = self.prepare_config_and_inputs()
UpperCamelCase_ = {'pixel_values': pixel_values, 'pixel_mask': pixel_mask}
return config, inputs_dict
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> List[Any]:
UpperCamelCase_ = output.encoder_hidden_states
UpperCamelCase_ = output.pixel_decoder_hidden_states
UpperCamelCase_ = output.transformer_decoder_hidden_states
self.parent.assertTrue(len(_UpperCAmelCase ) , len(config.backbone_config.depths ) )
self.parent.assertTrue(len(_UpperCAmelCase ) , len(config.backbone_config.depths ) )
self.parent.assertTrue(len(_UpperCAmelCase ) , config.decoder_layers )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase=False ) -> Any:
with torch.no_grad():
UpperCamelCase_ = MaskaFormerModel(config=_UpperCAmelCase )
model.to(_UpperCAmelCase )
model.eval()
UpperCamelCase_ = model(pixel_values=_UpperCAmelCase , pixel_mask=_UpperCAmelCase )
UpperCamelCase_ = model(_UpperCAmelCase , output_hidden_states=_UpperCAmelCase )
self.parent.assertEqual(
output.transformer_decoder_last_hidden_state.shape , (self.batch_size, self.num_queries, self.hidden_dim) , )
# let's ensure the other two hidden state exists
self.parent.assertTrue(output.pixel_decoder_last_hidden_state is not None )
self.parent.assertTrue(output.encoder_last_hidden_state is not None )
if output_hidden_states:
self.check_output_hidden_state(_UpperCAmelCase , _UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase ) -> Union[str, Any]:
UpperCamelCase_ = MaskaFormerForUniversalSegmentation(config=_UpperCAmelCase )
model.to(_UpperCAmelCase )
model.eval()
def comm_check_on_output(_UpperCAmelCase ):
# let's still check that all the required stuff is there
self.parent.assertTrue(result.transformer_decoder_last_hidden_state is not None )
self.parent.assertTrue(result.pixel_decoder_last_hidden_state is not None )
self.parent.assertTrue(result.encoder_last_hidden_state is not None )
# okay, now we need to check the logits shape
# due to the encoder compression, masks have a //4 spatial size
self.parent.assertEqual(
result.masks_queries_logits.shape , (self.batch_size, self.num_queries, self.min_size // 4, self.max_size // 4) , )
# + 1 for null class
self.parent.assertEqual(
result.class_queries_logits.shape , (self.batch_size, self.num_queries, self.num_labels + 1) )
with torch.no_grad():
UpperCamelCase_ = model(pixel_values=_UpperCAmelCase , pixel_mask=_UpperCAmelCase )
UpperCamelCase_ = model(_UpperCAmelCase )
comm_check_on_output(_UpperCAmelCase )
UpperCamelCase_ = model(
pixel_values=_UpperCAmelCase , pixel_mask=_UpperCAmelCase , mask_labels=_UpperCAmelCase , class_labels=_UpperCAmelCase )
comm_check_on_output(_UpperCAmelCase )
self.parent.assertTrue(result.loss is not None )
self.parent.assertEqual(result.loss.shape , torch.Size([1] ) )
@require_torch
class _a ( UpperCAmelCase__ , UpperCAmelCase__ , unittest.TestCase ):
"""simple docstring"""
A_ = (MaskaFormerModel, MaskaFormerForUniversalSegmentation) if is_torch_available() else ()
A_ = {"""feature-extraction""": MaskaFormerModel} if is_torch_available() else {}
A_ = False
A_ = False
A_ = False
A_ = False
def _UpperCAmelCase ( self ) -> Optional[Any]:
UpperCamelCase_ = MaskaFormerModelTester(self )
UpperCamelCase_ = ConfigTester(self , config_class=_UpperCAmelCase , has_text_modality=_UpperCAmelCase )
def _UpperCAmelCase ( self ) -> Union[str, Any]:
self.config_tester.run_common_tests()
def _UpperCAmelCase ( self ) -> Union[str, Any]:
UpperCamelCase_ , UpperCamelCase_ = self.model_tester.prepare_config_and_inputs_for_common()
self.model_tester.create_and_check_maskaformer_model(_UpperCAmelCase , **_UpperCAmelCase , output_hidden_states=_UpperCAmelCase )
def _UpperCAmelCase ( self ) -> List[Any]:
UpperCamelCase_ = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_maskaformer_instance_segmentation_head_model(*_UpperCAmelCase )
@unittest.skip(reason='Mask2Former does not use inputs_embeds' )
def _UpperCAmelCase ( self ) -> Any:
pass
@unittest.skip(reason='Mask2Former does not have a get_input_embeddings method' )
def _UpperCAmelCase ( self ) -> Optional[int]:
pass
@unittest.skip(reason='Mask2Former is not a generative model' )
def _UpperCAmelCase ( self ) -> Any:
pass
@unittest.skip(reason='Mask2Former does not use token embeddings' )
def _UpperCAmelCase ( self ) -> Optional[Any]:
pass
@require_torch_multi_gpu
@unittest.skip(
reason='Mask2Former has some layers using `add_module` which doesn\'t work well with `nn.DataParallel`' )
def _UpperCAmelCase ( self ) -> int:
pass
@unittest.skip('Will be fixed soon by reducing the size of the model used for common tests.' )
def _UpperCAmelCase ( self ) -> str:
pass
def _UpperCAmelCase ( self ) -> Optional[int]:
UpperCamelCase_ , UpperCamelCase_ = self.model_tester.prepare_config_and_inputs_for_common()
for model_class in self.all_model_classes:
UpperCamelCase_ = model_class(_UpperCAmelCase )
UpperCamelCase_ = inspect.signature(model.forward )
# signature.parameters is an OrderedDict => so arg_names order is deterministic
UpperCamelCase_ = [*signature.parameters.keys()]
UpperCamelCase_ = ['pixel_values']
self.assertListEqual(arg_names[:1] , _UpperCAmelCase )
@slow
def _UpperCAmelCase ( self ) -> Tuple:
for model_name in ["facebook/mask2former-swin-small-coco-instance"]:
UpperCamelCase_ = MaskaFormerModel.from_pretrained(_UpperCAmelCase )
self.assertIsNotNone(_UpperCAmelCase )
def _UpperCAmelCase ( self ) -> Dict:
UpperCamelCase_ = (self.model_tester.min_size,) * 2
UpperCamelCase_ = {
'pixel_values': torch.randn((2, 3, *size) , device=_UpperCAmelCase ),
'mask_labels': torch.randn((2, 10, *size) , device=_UpperCAmelCase ),
'class_labels': torch.zeros(2 , 10 , device=_UpperCAmelCase ).long(),
}
UpperCamelCase_ = self.model_tester.get_config()
UpperCamelCase_ = MaskaFormerForUniversalSegmentation(_UpperCAmelCase ).to(_UpperCAmelCase )
UpperCamelCase_ = model(**_UpperCAmelCase )
self.assertTrue(outputs.loss is not None )
def _UpperCAmelCase ( self ) -> str:
UpperCamelCase_ , UpperCamelCase_ = self.model_tester.prepare_config_and_inputs_for_common()
self.model_tester.create_and_check_maskaformer_model(_UpperCAmelCase , **_UpperCAmelCase , output_hidden_states=_UpperCAmelCase )
def _UpperCAmelCase ( self ) -> Optional[int]:
UpperCamelCase_ , UpperCamelCase_ = self.model_tester.prepare_config_and_inputs_for_common()
for model_class in self.all_model_classes:
UpperCamelCase_ = model_class(_UpperCAmelCase ).to(_UpperCAmelCase )
UpperCamelCase_ = model(**_UpperCAmelCase , output_attentions=_UpperCAmelCase )
self.assertTrue(outputs.attentions is not None )
def _UpperCAmelCase ( self ) -> List[Any]:
if not self.model_tester.is_training:
return
UpperCamelCase_ = self.all_model_classes[1]
UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ = self.model_tester.prepare_config_and_inputs()
UpperCamelCase_ = model_class(_UpperCAmelCase )
model.to(_UpperCAmelCase )
model.train()
UpperCamelCase_ = model(_UpperCAmelCase , mask_labels=_UpperCAmelCase , class_labels=_UpperCAmelCase ).loss
loss.backward()
def _UpperCAmelCase ( self ) -> int:
UpperCamelCase_ = self.all_model_classes[1]
UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ = self.model_tester.prepare_config_and_inputs()
UpperCamelCase_ = True
UpperCamelCase_ = True
UpperCamelCase_ = model_class(_UpperCAmelCase ).to(_UpperCAmelCase )
model.train()
UpperCamelCase_ = model(_UpperCAmelCase , mask_labels=_UpperCAmelCase , class_labels=_UpperCAmelCase )
UpperCamelCase_ = outputs.encoder_hidden_states[0]
encoder_hidden_states.retain_grad()
UpperCamelCase_ = outputs.pixel_decoder_hidden_states[0]
pixel_decoder_hidden_states.retain_grad()
UpperCamelCase_ = outputs.transformer_decoder_hidden_states[0]
transformer_decoder_hidden_states.retain_grad()
UpperCamelCase_ = outputs.attentions[0]
attentions.retain_grad()
outputs.loss.backward(retain_graph=_UpperCAmelCase )
self.assertIsNotNone(encoder_hidden_states.grad )
self.assertIsNotNone(pixel_decoder_hidden_states.grad )
self.assertIsNotNone(transformer_decoder_hidden_states.grad )
self.assertIsNotNone(attentions.grad )
snake_case__ : List[Any] = 1E-4
def _snake_case ():
UpperCamelCase_ = Image.open('./tests/fixtures/tests_samples/COCO/000000039769.png')
return image
@require_vision
@slow
class _a ( unittest.TestCase ):
"""simple docstring"""
@cached_property
def _UpperCAmelCase ( self ) -> Optional[int]:
return "facebook/mask2former-swin-small-coco-instance"
@cached_property
def _UpperCAmelCase ( self ) -> List[str]:
return MaskaFormerImageProcessor.from_pretrained(self.model_checkpoints ) if is_vision_available() else None
def _UpperCAmelCase ( self ) -> str:
UpperCamelCase_ = MaskaFormerModel.from_pretrained(self.model_checkpoints ).to(_UpperCAmelCase )
UpperCamelCase_ = self.default_image_processor
UpperCamelCase_ = prepare_img()
UpperCamelCase_ = image_processor(_UpperCAmelCase , return_tensors='pt' ).to(_UpperCAmelCase )
UpperCamelCase_ = inputs['pixel_values'].shape
# check size is divisible by 32
self.assertTrue((inputs_shape[-1] % 32) == 0 and (inputs_shape[-2] % 32) == 0 )
# check size
self.assertEqual(_UpperCAmelCase , (1, 3, 384, 384) )
with torch.no_grad():
UpperCamelCase_ = model(**_UpperCAmelCase )
UpperCamelCase_ = torch.tensor(
[[-0.2_7_9_0, -1.0_7_1_7, -1.1_6_6_8], [-0.5_1_2_8, -0.3_1_2_8, -0.4_9_8_7], [-0.5_8_3_2, 0.1_9_7_1, -0.0_1_9_7]] ).to(_UpperCAmelCase )
self.assertTrue(
torch.allclose(
outputs.encoder_last_hidden_state[0, 0, :3, :3] , _UpperCAmelCase , atol=_UpperCAmelCase ) )
UpperCamelCase_ = torch.tensor(
[[0.8_9_7_3, 1.1_8_4_7, 1.1_7_7_6], [1.1_9_3_4, 1.5_0_4_0, 1.5_1_2_8], [1.1_1_5_3, 1.4_4_8_6, 1.4_9_5_1]] ).to(_UpperCAmelCase )
self.assertTrue(
torch.allclose(
outputs.pixel_decoder_last_hidden_state[0, 0, :3, :3] , _UpperCAmelCase , atol=_UpperCAmelCase ) )
UpperCamelCase_ = torch.tensor(
[[2.1_1_5_2, 1.7_0_0_0, -0.8_6_0_3], [1.5_8_0_8, 1.8_0_0_4, -0.9_3_5_3], [1.6_0_4_3, 1.7_4_9_5, -0.5_9_9_9]] ).to(_UpperCAmelCase )
self.assertTrue(
torch.allclose(
outputs.transformer_decoder_last_hidden_state[0, :3, :3] , _UpperCAmelCase , atol=_UpperCAmelCase ) )
def _UpperCAmelCase ( self ) -> Optional[Any]:
UpperCamelCase_ = MaskaFormerForUniversalSegmentation.from_pretrained(self.model_checkpoints ).to(_UpperCAmelCase ).eval()
UpperCamelCase_ = self.default_image_processor
UpperCamelCase_ = prepare_img()
UpperCamelCase_ = image_processor(_UpperCAmelCase , return_tensors='pt' ).to(_UpperCAmelCase )
UpperCamelCase_ = inputs['pixel_values'].shape
# check size is divisible by 32
self.assertTrue((inputs_shape[-1] % 32) == 0 and (inputs_shape[-2] % 32) == 0 )
# check size
self.assertEqual(_UpperCAmelCase , (1, 3, 384, 384) )
with torch.no_grad():
UpperCamelCase_ = model(**_UpperCAmelCase )
# masks_queries_logits
UpperCamelCase_ = outputs.masks_queries_logits
self.assertEqual(
masks_queries_logits.shape , (1, model.config.num_queries, inputs_shape[-2] // 4, inputs_shape[-1] // 4) )
UpperCamelCase_ = [
[-8.7_8_3_9, -9.0_0_5_6, -8.8_1_2_1],
[-7.4_1_0_4, -7.0_3_1_3, -6.5_4_0_1],
[-6.6_1_0_5, -6.3_4_2_7, -6.4_6_7_5],
]
UpperCamelCase_ = torch.tensor(_UpperCAmelCase ).to(_UpperCAmelCase )
self.assertTrue(torch.allclose(masks_queries_logits[0, 0, :3, :3] , _UpperCAmelCase , atol=_UpperCAmelCase ) )
# class_queries_logits
UpperCamelCase_ = outputs.class_queries_logits
self.assertEqual(class_queries_logits.shape , (1, model.config.num_queries, model.config.num_labels + 1) )
UpperCamelCase_ = torch.tensor(
[
[1.8_3_2_4, -8.0_8_3_5, -4.1_9_2_2],
[0.8_4_5_0, -9.0_0_5_0, -3.6_0_5_3],
[0.3_0_4_5, -7.7_2_9_3, -3.0_2_7_5],
] ).to(_UpperCAmelCase )
self.assertTrue(torch.allclose(outputs.class_queries_logits[0, :3, :3] , _UpperCAmelCase , atol=_UpperCAmelCase ) )
def _UpperCAmelCase ( self ) -> Dict:
UpperCamelCase_ = MaskaFormerForUniversalSegmentation.from_pretrained(self.model_checkpoints ).to(_UpperCAmelCase ).eval()
UpperCamelCase_ = self.default_image_processor
UpperCamelCase_ = image_processor(
[np.zeros((3, 800, 1333) ), np.zeros((3, 800, 1333) )] , segmentation_maps=[np.zeros((384, 384) ).astype(np.floataa ), np.zeros((384, 384) ).astype(np.floataa )] , return_tensors='pt' , )
UpperCamelCase_ = inputs['pixel_values'].to(_UpperCAmelCase )
UpperCamelCase_ = [el.to(_UpperCAmelCase ) for el in inputs['mask_labels']]
UpperCamelCase_ = [el.to(_UpperCAmelCase ) for el in inputs['class_labels']]
with torch.no_grad():
UpperCamelCase_ = model(**_UpperCAmelCase )
self.assertTrue(outputs.loss is not None )
| 23 | 1 |
from collections import OrderedDict
from typing import Mapping
from ...configuration_utils import PretrainedConfig
from ...onnx import OnnxConfig
from ...utils import logging
snake_case__ : str = logging.get_logger(__name__)
snake_case__ : List[str] = {
"""xlm-roberta-base""": """https://huggingface.co/xlm-roberta-base/resolve/main/config.json""",
"""xlm-roberta-large""": """https://huggingface.co/xlm-roberta-large/resolve/main/config.json""",
"""xlm-roberta-large-finetuned-conll02-dutch""": (
"""https://huggingface.co/xlm-roberta-large-finetuned-conll02-dutch/resolve/main/config.json"""
),
"""xlm-roberta-large-finetuned-conll02-spanish""": (
"""https://huggingface.co/xlm-roberta-large-finetuned-conll02-spanish/resolve/main/config.json"""
),
"""xlm-roberta-large-finetuned-conll03-english""": (
"""https://huggingface.co/xlm-roberta-large-finetuned-conll03-english/resolve/main/config.json"""
),
"""xlm-roberta-large-finetuned-conll03-german""": (
"""https://huggingface.co/xlm-roberta-large-finetuned-conll03-german/resolve/main/config.json"""
),
}
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
A_ = """xlm-roberta"""
def __init__( self , _UpperCAmelCase=30522 , _UpperCAmelCase=768 , _UpperCAmelCase=12 , _UpperCAmelCase=12 , _UpperCAmelCase=3072 , _UpperCAmelCase="gelu" , _UpperCAmelCase=0.1 , _UpperCAmelCase=0.1 , _UpperCAmelCase=512 , _UpperCAmelCase=2 , _UpperCAmelCase=0.0_2 , _UpperCAmelCase=1e-12 , _UpperCAmelCase=1 , _UpperCAmelCase=0 , _UpperCAmelCase=2 , _UpperCAmelCase="absolute" , _UpperCAmelCase=True , _UpperCAmelCase=None , **_UpperCAmelCase , ) -> List[str]:
super().__init__(pad_token_id=_UpperCAmelCase , bos_token_id=_UpperCAmelCase , eos_token_id=_UpperCAmelCase , **_UpperCAmelCase )
UpperCamelCase_ = vocab_size
UpperCamelCase_ = hidden_size
UpperCamelCase_ = num_hidden_layers
UpperCamelCase_ = num_attention_heads
UpperCamelCase_ = hidden_act
UpperCamelCase_ = intermediate_size
UpperCamelCase_ = hidden_dropout_prob
UpperCamelCase_ = attention_probs_dropout_prob
UpperCamelCase_ = max_position_embeddings
UpperCamelCase_ = type_vocab_size
UpperCamelCase_ = initializer_range
UpperCamelCase_ = layer_norm_eps
UpperCamelCase_ = position_embedding_type
UpperCamelCase_ = use_cache
UpperCamelCase_ = classifier_dropout
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
@property
def _UpperCAmelCase ( self ) -> Mapping[str, Mapping[int, str]]:
if self.task == "multiple-choice":
UpperCamelCase_ = {0: 'batch', 1: 'choice', 2: 'sequence'}
else:
UpperCamelCase_ = {0: 'batch', 1: 'sequence'}
return OrderedDict(
[
('input_ids', dynamic_axis),
('attention_mask', dynamic_axis),
] )
| 23 |
import copy
from typing import TYPE_CHECKING, Any, Mapping, Optional, OrderedDict
from packaging import version
from ...configuration_utils import PretrainedConfig
from ...onnx import OnnxConfig
from ...utils import logging
from ..auto.configuration_auto import AutoConfig
if TYPE_CHECKING:
from ... import PreTrainedTokenizerBase, TensorType
snake_case__ : List[str] = logging.get_logger(__name__)
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
A_ = """vision-encoder-decoder"""
A_ = True
def __init__( self , **_UpperCAmelCase ) -> Dict:
super().__init__(**_UpperCAmelCase )
if "encoder" not in kwargs or "decoder" not in kwargs:
raise ValueError(
f"""A configuraton of type {self.model_type} cannot be instantiated because """
f"""not both `encoder` and `decoder` sub-configurations are passed, but only {kwargs}""" )
UpperCamelCase_ = kwargs.pop('encoder' )
UpperCamelCase_ = encoder_config.pop('model_type' )
UpperCamelCase_ = kwargs.pop('decoder' )
UpperCamelCase_ = decoder_config.pop('model_type' )
UpperCamelCase_ = AutoConfig.for_model(_UpperCAmelCase , **_UpperCAmelCase )
UpperCamelCase_ = AutoConfig.for_model(_UpperCAmelCase , **_UpperCAmelCase )
UpperCamelCase_ = True
@classmethod
def _UpperCAmelCase ( cls , _UpperCAmelCase , _UpperCAmelCase , **_UpperCAmelCase ) -> PretrainedConfig:
logger.info('Setting `config.is_decoder=True` and `config.add_cross_attention=True` for decoder_config' )
UpperCamelCase_ = True
UpperCamelCase_ = True
return cls(encoder=encoder_config.to_dict() , decoder=decoder_config.to_dict() , **_UpperCAmelCase )
def _UpperCAmelCase ( self ) -> int:
UpperCamelCase_ = copy.deepcopy(self.__dict__ )
UpperCamelCase_ = self.encoder.to_dict()
UpperCamelCase_ = self.decoder.to_dict()
UpperCamelCase_ = self.__class__.model_type
return output
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
A_ = version.parse("""1.11""" )
@property
def _UpperCAmelCase ( self ) -> Mapping[str, Mapping[int, str]]:
return OrderedDict(
[
('pixel_values', {0: 'batch', 1: 'num_channels', 2: 'height', 3: 'width'}),
] )
@property
def _UpperCAmelCase ( self ) -> float:
return 1e-4
@property
def _UpperCAmelCase ( self ) -> Mapping[str, Mapping[int, str]]:
return OrderedDict({'last_hidden_state': {0: 'batch', 1: 'encoder_sequence'}} )
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
@property
def _UpperCAmelCase ( self ) -> Mapping[str, Mapping[int, str]]:
UpperCamelCase_ = OrderedDict()
UpperCamelCase_ = {0: 'batch', 1: 'past_decoder_sequence + sequence'}
UpperCamelCase_ = {0: 'batch', 1: 'past_decoder_sequence + sequence'}
UpperCamelCase_ = {0: 'batch', 1: 'encoder_sequence'}
return common_inputs
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase = -1 , _UpperCAmelCase = -1 , _UpperCAmelCase = False , _UpperCAmelCase = None , ) -> Mapping[str, Any]:
import torch
UpperCamelCase_ = OrderedDict()
UpperCamelCase_ = super().generate_dummy_inputs(
_UpperCAmelCase , batch_size=_UpperCAmelCase , seq_length=_UpperCAmelCase , is_pair=_UpperCAmelCase , framework=_UpperCAmelCase )
UpperCamelCase_ , UpperCamelCase_ = dummy_input['input_ids'].shape
UpperCamelCase_ = (batch, encoder_sequence, self._config.encoder_hidden_size)
UpperCamelCase_ = dummy_input.pop('input_ids' )
UpperCamelCase_ = dummy_input.pop('attention_mask' )
UpperCamelCase_ = torch.zeros(_UpperCAmelCase )
return common_inputs
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
@property
def _UpperCAmelCase ( self ) -> None:
pass
def _UpperCAmelCase ( self , _UpperCAmelCase ) -> OnnxConfig:
return VisionEncoderDecoderEncoderOnnxConfig(_UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase = "default" ) -> OnnxConfig:
UpperCamelCase_ = encoder_config.hidden_size
return VisionEncoderDecoderDecoderOnnxConfig(_UpperCAmelCase , _UpperCAmelCase )
| 23 | 1 |
from typing import TYPE_CHECKING
from ...utils import OptionalDependencyNotAvailable, _LazyModule, is_flax_available, is_torch_available
SCREAMING_SNAKE_CASE__ : Dict = {
"""configuration_gpt_neo""": ["""GPT_NEO_PRETRAINED_CONFIG_ARCHIVE_MAP""", """GPTNeoConfig""", """GPTNeoOnnxConfig"""],
}
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
SCREAMING_SNAKE_CASE__ : Dict = [
"""GPT_NEO_PRETRAINED_MODEL_ARCHIVE_LIST""",
"""GPTNeoForCausalLM""",
"""GPTNeoForQuestionAnswering""",
"""GPTNeoForSequenceClassification""",
"""GPTNeoForTokenClassification""",
"""GPTNeoModel""",
"""GPTNeoPreTrainedModel""",
"""load_tf_weights_in_gpt_neo""",
]
try:
if not is_flax_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
SCREAMING_SNAKE_CASE__ : Union[str, Any] = [
"""FlaxGPTNeoForCausalLM""",
"""FlaxGPTNeoModel""",
"""FlaxGPTNeoPreTrainedModel""",
]
if TYPE_CHECKING:
from .configuration_gpt_neo import GPT_NEO_PRETRAINED_CONFIG_ARCHIVE_MAP, GPTNeoConfig, GPTNeoOnnxConfig
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_gpt_neo import (
GPT_NEO_PRETRAINED_MODEL_ARCHIVE_LIST,
GPTNeoForCausalLM,
GPTNeoForQuestionAnswering,
GPTNeoForSequenceClassification,
GPTNeoForTokenClassification,
GPTNeoModel,
GPTNeoPreTrainedModel,
load_tf_weights_in_gpt_neo,
)
try:
if not is_flax_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_flax_gpt_neo import FlaxGPTNeoForCausalLM, FlaxGPTNeoModel, FlaxGPTNeoPreTrainedModel
else:
import sys
SCREAMING_SNAKE_CASE__ : List[str] = _LazyModule(__name__, globals()["""__file__"""], _import_structure, module_spec=__spec__)
| 0 |
import argparse
import torch
from transformers import MobileBertConfig, MobileBertForPreTraining, load_tf_weights_in_mobilebert
from transformers.utils import logging
logging.set_verbosity_info()
def _snake_case (__lowercase , __lowercase , __lowercase):
# Initialise PyTorch model
UpperCamelCase_ = MobileBertConfig.from_json_file(__lowercase)
print(f"""Building PyTorch model from configuration: {config}""")
UpperCamelCase_ = MobileBertForPreTraining(__lowercase)
# Load weights from tf checkpoint
UpperCamelCase_ = load_tf_weights_in_mobilebert(__lowercase , __lowercase , __lowercase)
# Save pytorch-model
print(f"""Save PyTorch model to {pytorch_dump_path}""")
torch.save(model.state_dict() , __lowercase)
if __name__ == "__main__":
snake_case__ : int = argparse.ArgumentParser()
# Required parameters
parser.add_argument(
"""--tf_checkpoint_path""", default=None, type=str, required=True, help="""Path to the TensorFlow checkpoint path."""
)
parser.add_argument(
"""--mobilebert_config_file""",
default=None,
type=str,
required=True,
help=(
"""The config json file corresponding to the pre-trained MobileBERT model. \n"""
"""This specifies the model architecture."""
),
)
parser.add_argument(
"""--pytorch_dump_path""", default=None, type=str, required=True, help="""Path to the output PyTorch model."""
)
snake_case__ : Optional[Any] = parser.parse_args()
convert_tf_checkpoint_to_pytorch(args.tf_checkpoint_path, args.mobilebert_config_file, args.pytorch_dump_path)
| 23 | 0 |
from typing import Any
class __lowerCamelCase :
def __init__( self: int,A_: Any ):
'''simple docstring'''
__UpperCamelCase = data
__UpperCamelCase = None
def __repr__( self: Any ):
'''simple docstring'''
return F'''Node({self.data})'''
class __lowerCamelCase :
def __init__( self: Union[str, Any] ):
'''simple docstring'''
__UpperCamelCase = None
def __iter__( self: int ):
'''simple docstring'''
__UpperCamelCase = self.head
while node:
yield node.data
__UpperCamelCase = node.next
def __len__( self: List[str] ):
'''simple docstring'''
return sum(1 for _ in self )
def __repr__( self: Any ):
'''simple docstring'''
return "->".join([str(A_ ) for item in self] )
def __getitem__( self: int,A_: int ):
'''simple docstring'''
if not 0 <= index < len(self ):
raise ValueError('list index out of range.' )
for i, node in enumerate(self ):
if i == index:
return node
return None
def __setitem__( self: int,A_: int,A_: Any ):
'''simple docstring'''
if not 0 <= index < len(self ):
raise ValueError('list index out of range.' )
__UpperCamelCase = self.head
for _ in range(A_ ):
__UpperCamelCase = current.next
__UpperCamelCase = data
def snake_case_ ( self: Union[str, Any],A_: Any ):
'''simple docstring'''
self.insert_nth(len(self ),A_ )
def snake_case_ ( self: List[Any],A_: Any ):
'''simple docstring'''
self.insert_nth(0,A_ )
def snake_case_ ( self: Optional[Any],A_: int,A_: Any ):
'''simple docstring'''
if not 0 <= index <= len(self ):
raise IndexError('list index out of range' )
__UpperCamelCase = Node(A_ )
if self.head is None:
__UpperCamelCase = new_node
elif index == 0:
__UpperCamelCase = self.head # link new_node to head
__UpperCamelCase = new_node
else:
__UpperCamelCase = self.head
for _ in range(index - 1 ):
__UpperCamelCase = temp.next
__UpperCamelCase = temp.next
__UpperCamelCase = new_node
def snake_case_ ( self: str ): # print every node data
'''simple docstring'''
print(self )
def snake_case_ ( self: int ):
'''simple docstring'''
return self.delete_nth(0 )
def snake_case_ ( self: str ): # delete from tail
'''simple docstring'''
return self.delete_nth(len(self ) - 1 )
def snake_case_ ( self: Any,A_: int = 0 ):
'''simple docstring'''
if not 0 <= index <= len(self ) - 1: # test if index is valid
raise IndexError('List index out of range.' )
__UpperCamelCase = self.head # default first node
if index == 0:
__UpperCamelCase = self.head.next
else:
__UpperCamelCase = self.head
for _ in range(index - 1 ):
__UpperCamelCase = temp.next
__UpperCamelCase = temp.next
__UpperCamelCase = temp.next.next
return delete_node.data
def snake_case_ ( self: Any ):
'''simple docstring'''
return self.head is None
def snake_case_ ( self: Optional[int] ):
'''simple docstring'''
__UpperCamelCase = None
__UpperCamelCase = self.head
while current:
# Store the current node's next node.
__UpperCamelCase = current.next
# Make the current node's next point backwards
__UpperCamelCase = prev
# Make the previous node be the current node
__UpperCamelCase = current
# Make the current node the next node (to progress iteration)
__UpperCamelCase = next_node
# Return prev in order to put the head at the end
__UpperCamelCase = prev
def _A ( ) -> None:
"""simple docstring"""
__UpperCamelCase = LinkedList()
assert linked_list.is_empty() is True
assert str(_lowercase ) == ""
try:
linked_list.delete_head()
raise AssertionError # This should not happen.
except IndexError:
assert True # This should happen.
try:
linked_list.delete_tail()
raise AssertionError # This should not happen.
except IndexError:
assert True # This should happen.
for i in range(10 ):
assert len(_lowercase ) == i
linked_list.insert_nth(_lowercase , i + 1 )
assert str(_lowercase ) == "->".join(str(_lowercase ) for i in range(1 , 11 ) )
linked_list.insert_head(0 )
linked_list.insert_tail(11 )
assert str(_lowercase ) == "->".join(str(_lowercase ) for i in range(0 , 12 ) )
assert linked_list.delete_head() == 0
assert linked_list.delete_nth(9 ) == 10
assert linked_list.delete_tail() == 11
assert len(_lowercase ) == 9
assert str(_lowercase ) == "->".join(str(_lowercase ) for i in range(1 , 10 ) )
assert all(linked_list[i] == i + 1 for i in range(0 , 9 ) ) is True
for i in range(0 , 9 ):
__UpperCamelCase = -i
assert all(linked_list[i] == -i for i in range(0 , 9 ) ) is True
linked_list.reverse()
assert str(_lowercase ) == "->".join(str(_lowercase ) for i in range(-8 , 1 ) )
def _A ( ) -> None:
"""simple docstring"""
__UpperCamelCase = [
-9,
1_00,
Node(77_34_51_12 ),
'dlrow olleH',
7,
55_55,
0,
-1_92.5_55_55,
'Hello, world!',
77.9,
Node(10 ),
None,
None,
12.20,
]
__UpperCamelCase = LinkedList()
for i in test_input:
linked_list.insert_tail(_lowercase )
# Check if it's empty or not
assert linked_list.is_empty() is False
assert (
str(_lowercase ) == "-9->100->Node(77345112)->dlrow olleH->7->5555->0->"
"-192.55555->Hello, world!->77.9->Node(10)->None->None->12.2"
)
# Delete the head
__UpperCamelCase = linked_list.delete_head()
assert result == -9
assert (
str(_lowercase ) == "100->Node(77345112)->dlrow olleH->7->5555->0->-192.55555->"
"Hello, world!->77.9->Node(10)->None->None->12.2"
)
# Delete the tail
__UpperCamelCase = linked_list.delete_tail()
assert result == 12.2
assert (
str(_lowercase ) == "100->Node(77345112)->dlrow olleH->7->5555->0->-192.55555->"
"Hello, world!->77.9->Node(10)->None->None"
)
# Delete a node in specific location in linked list
__UpperCamelCase = linked_list.delete_nth(10 )
assert result is None
assert (
str(_lowercase ) == "100->Node(77345112)->dlrow olleH->7->5555->0->-192.55555->"
"Hello, world!->77.9->Node(10)->None"
)
# Add a Node instance to its head
linked_list.insert_head(Node('Hello again, world!' ) )
assert (
str(_lowercase )
== "Node(Hello again, world!)->100->Node(77345112)->dlrow olleH->"
"7->5555->0->-192.55555->Hello, world!->77.9->Node(10)->None"
)
# Add None to its tail
linked_list.insert_tail(_lowercase )
assert (
str(_lowercase )
== "Node(Hello again, world!)->100->Node(77345112)->dlrow olleH->"
"7->5555->0->-192.55555->Hello, world!->77.9->Node(10)->None->None"
)
# Reverse the linked list
linked_list.reverse()
assert (
str(_lowercase )
== "None->None->Node(10)->77.9->Hello, world!->-192.55555->0->5555->"
"7->dlrow olleH->Node(77345112)->100->Node(Hello again, world!)"
)
def _A ( ) -> List[str]:
"""simple docstring"""
from doctest import testmod
testmod()
__UpperCamelCase = LinkedList()
linked_list.insert_head(input('Inserting 1st at head ' ).strip() )
linked_list.insert_head(input('Inserting 2nd at head ' ).strip() )
print('\nPrint list:' )
linked_list.print_list()
linked_list.insert_tail(input('\nInserting 1st at tail ' ).strip() )
linked_list.insert_tail(input('Inserting 2nd at tail ' ).strip() )
print('\nPrint list:' )
linked_list.print_list()
print('\nDelete head' )
linked_list.delete_head()
print('Delete tail' )
linked_list.delete_tail()
print('\nPrint list:' )
linked_list.print_list()
print('\nReverse linked list' )
linked_list.reverse()
print('\nPrint list:' )
linked_list.print_list()
print('\nString representation of linked list:' )
print(_lowercase )
print('\nReading/changing Node data using indexing:' )
print(f'''Element at Position 1: {linked_list[1]}''' )
__UpperCamelCase = input('Enter New Value: ' ).strip()
print('New list:' )
print(_lowercase )
print(f'''length of linked_list is : {len(_lowercase )}''' )
if __name__ == "__main__":
main()
| 1 |
import gc
import unittest
from transformers import MODEL_FOR_MASKED_LM_MAPPING, TF_MODEL_FOR_MASKED_LM_MAPPING, FillMaskPipeline, pipeline
from transformers.pipelines import PipelineException
from transformers.testing_utils import (
is_pipeline_test,
is_torch_available,
nested_simplify,
require_tf,
require_torch,
require_torch_gpu,
slow,
)
from .test_pipelines_common import ANY
@is_pipeline_test
class _a ( unittest.TestCase ):
"""simple docstring"""
A_ = MODEL_FOR_MASKED_LM_MAPPING
A_ = TF_MODEL_FOR_MASKED_LM_MAPPING
def _UpperCAmelCase ( self ) -> List[str]:
super().tearDown()
# clean-up as much as possible GPU memory occupied by PyTorch
gc.collect()
if is_torch_available():
import torch
torch.cuda.empty_cache()
@require_tf
def _UpperCAmelCase ( self ) -> str:
UpperCamelCase_ = pipeline(task='fill-mask' , model='sshleifer/tiny-distilroberta-base' , top_k=2 , framework='tf' )
UpperCamelCase_ = unmasker('My name is <mask>' )
self.assertEqual(
nested_simplify(_UpperCAmelCase , decimals=6 ) , [
{'sequence': 'My name is grouped', 'score': 2.1e-05, 'token': 38015, 'token_str': ' grouped'},
{'sequence': 'My name is accuser', 'score': 2.1e-05, 'token': 25506, 'token_str': ' accuser'},
] , )
UpperCamelCase_ = unmasker('The largest city in France is <mask>' )
self.assertEqual(
nested_simplify(_UpperCAmelCase , decimals=6 ) , [
{
'sequence': 'The largest city in France is grouped',
'score': 2.1e-05,
'token': 38015,
'token_str': ' grouped',
},
{
'sequence': 'The largest city in France is accuser',
'score': 2.1e-05,
'token': 25506,
'token_str': ' accuser',
},
] , )
UpperCamelCase_ = unmasker('My name is <mask>' , targets=[' Patrick', ' Clara', ' Teven'] , top_k=3 )
self.assertEqual(
nested_simplify(_UpperCAmelCase , decimals=6 ) , [
{'sequence': 'My name is Clara', 'score': 2e-05, 'token': 13606, 'token_str': ' Clara'},
{'sequence': 'My name is Patrick', 'score': 2e-05, 'token': 3499, 'token_str': ' Patrick'},
{'sequence': 'My name is Te', 'score': 1.9e-05, 'token': 2941, 'token_str': ' Te'},
] , )
@require_torch
def _UpperCAmelCase ( self ) -> Union[str, Any]:
UpperCamelCase_ = pipeline(task='fill-mask' , model='sshleifer/tiny-distilroberta-base' , top_k=2 , framework='pt' )
UpperCamelCase_ = unmasker('My name is <mask>' )
self.assertEqual(
nested_simplify(_UpperCAmelCase , decimals=6 ) , [
{'sequence': 'My name is Maul', 'score': 2.2e-05, 'token': 35676, 'token_str': ' Maul'},
{'sequence': 'My name isELS', 'score': 2.2e-05, 'token': 16416, 'token_str': 'ELS'},
] , )
UpperCamelCase_ = unmasker('The largest city in France is <mask>' )
self.assertEqual(
nested_simplify(_UpperCAmelCase , decimals=6 ) , [
{
'sequence': 'The largest city in France is Maul',
'score': 2.2e-05,
'token': 35676,
'token_str': ' Maul',
},
{'sequence': 'The largest city in France isELS', 'score': 2.2e-05, 'token': 16416, 'token_str': 'ELS'},
] , )
UpperCamelCase_ = unmasker('My name is <mask>' , targets=[' Patrick', ' Clara', ' Teven'] , top_k=3 )
self.assertEqual(
nested_simplify(_UpperCAmelCase , decimals=6 ) , [
{'sequence': 'My name is Patrick', 'score': 2.1e-05, 'token': 3499, 'token_str': ' Patrick'},
{'sequence': 'My name is Te', 'score': 2e-05, 'token': 2941, 'token_str': ' Te'},
{'sequence': 'My name is Clara', 'score': 2e-05, 'token': 13606, 'token_str': ' Clara'},
] , )
UpperCamelCase_ = unmasker('My name is <mask> <mask>' , top_k=2 )
self.assertEqual(
nested_simplify(_UpperCAmelCase , decimals=6 ) , [
[
{
'score': 2.2e-05,
'token': 35676,
'token_str': ' Maul',
'sequence': '<s>My name is Maul<mask></s>',
},
{'score': 2.2e-05, 'token': 16416, 'token_str': 'ELS', 'sequence': '<s>My name isELS<mask></s>'},
],
[
{
'score': 2.2e-05,
'token': 35676,
'token_str': ' Maul',
'sequence': '<s>My name is<mask> Maul</s>',
},
{'score': 2.2e-05, 'token': 16416, 'token_str': 'ELS', 'sequence': '<s>My name is<mask>ELS</s>'},
],
] , )
@require_torch_gpu
def _UpperCAmelCase ( self ) -> Optional[Any]:
UpperCamelCase_ = pipeline('fill-mask' , model='hf-internal-testing/tiny-random-distilbert' , device=0 , framework='pt' )
# convert model to fp16
pipe.model.half()
UpperCamelCase_ = pipe('Paris is the [MASK] of France.' )
# We actually don't care about the result, we just want to make sure
# it works, meaning the float16 tensor got casted back to float32
# for postprocessing.
self.assertIsInstance(_UpperCAmelCase , _UpperCAmelCase )
@slow
@require_torch
def _UpperCAmelCase ( self ) -> List[Any]:
UpperCamelCase_ = pipeline(task='fill-mask' , model='distilroberta-base' , top_k=2 , framework='pt' )
self.run_large_test(_UpperCAmelCase )
@slow
@require_tf
def _UpperCAmelCase ( self ) -> Union[str, Any]:
UpperCamelCase_ = pipeline(task='fill-mask' , model='distilroberta-base' , top_k=2 , framework='tf' )
self.run_large_test(_UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase ) -> Tuple:
UpperCamelCase_ = unmasker('My name is <mask>' )
self.assertEqual(
nested_simplify(_UpperCAmelCase ) , [
{'sequence': 'My name is John', 'score': 0.0_0_8, 'token': 610, 'token_str': ' John'},
{'sequence': 'My name is Chris', 'score': 0.0_0_7, 'token': 1573, 'token_str': ' Chris'},
] , )
UpperCamelCase_ = unmasker('The largest city in France is <mask>' )
self.assertEqual(
nested_simplify(_UpperCAmelCase ) , [
{
'sequence': 'The largest city in France is Paris',
'score': 0.2_5_1,
'token': 2201,
'token_str': ' Paris',
},
{
'sequence': 'The largest city in France is Lyon',
'score': 0.2_1_4,
'token': 12790,
'token_str': ' Lyon',
},
] , )
UpperCamelCase_ = unmasker('My name is <mask>' , targets=[' Patrick', ' Clara', ' Teven'] , top_k=3 )
self.assertEqual(
nested_simplify(_UpperCAmelCase ) , [
{'sequence': 'My name is Patrick', 'score': 0.0_0_5, 'token': 3499, 'token_str': ' Patrick'},
{'sequence': 'My name is Clara', 'score': 0.0_0_0, 'token': 13606, 'token_str': ' Clara'},
{'sequence': 'My name is Te', 'score': 0.0_0_0, 'token': 2941, 'token_str': ' Te'},
] , )
@require_torch
def _UpperCAmelCase ( self ) -> List[Any]:
UpperCamelCase_ = pipeline(task='fill-mask' , model='sshleifer/tiny-distilroberta-base' , framework='pt' )
UpperCamelCase_ = None
UpperCamelCase_ = None
self.run_pipeline_test(_UpperCAmelCase , [] )
@require_tf
def _UpperCAmelCase ( self ) -> Optional[int]:
UpperCamelCase_ = pipeline(task='fill-mask' , model='sshleifer/tiny-distilroberta-base' , framework='tf' )
UpperCamelCase_ = None
UpperCamelCase_ = None
self.run_pipeline_test(_UpperCAmelCase , [] )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase ) -> Optional[Any]:
if tokenizer is None or tokenizer.mask_token_id is None:
self.skipTest('The provided tokenizer has no mask token, (probably reformer or wav2vec2)' )
UpperCamelCase_ = FillMaskPipeline(model=_UpperCAmelCase , tokenizer=_UpperCAmelCase )
UpperCamelCase_ = [
f"""This is another {tokenizer.mask_token} test""",
]
return fill_masker, examples
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> Union[str, Any]:
UpperCamelCase_ = fill_masker.tokenizer
UpperCamelCase_ = fill_masker.model
UpperCamelCase_ = fill_masker(
f"""This is a {tokenizer.mask_token}""" , )
self.assertEqual(
_UpperCAmelCase , [
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
] , )
UpperCamelCase_ = fill_masker([f"""This is a {tokenizer.mask_token}"""] )
self.assertEqual(
_UpperCAmelCase , [
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
] , )
UpperCamelCase_ = fill_masker([f"""This is a {tokenizer.mask_token}""", f"""Another {tokenizer.mask_token} great test."""] )
self.assertEqual(
_UpperCAmelCase , [
[
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
],
[
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
],
] , )
with self.assertRaises(_UpperCAmelCase ):
fill_masker([None] )
# No mask_token is not supported
with self.assertRaises(_UpperCAmelCase ):
fill_masker('This is' )
self.run_test_top_k(_UpperCAmelCase , _UpperCAmelCase )
self.run_test_targets(_UpperCAmelCase , _UpperCAmelCase )
self.run_test_top_k_targets(_UpperCAmelCase , _UpperCAmelCase )
self.fill_mask_with_duplicate_targets_and_top_k(_UpperCAmelCase , _UpperCAmelCase )
self.fill_mask_with_multiple_masks(_UpperCAmelCase , _UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> Optional[Any]:
UpperCamelCase_ = tokenizer.get_vocab()
UpperCamelCase_ = sorted(vocab.keys() )[:2]
# Pipeline argument
UpperCamelCase_ = FillMaskPipeline(model=_UpperCAmelCase , tokenizer=_UpperCAmelCase , targets=_UpperCAmelCase )
UpperCamelCase_ = fill_masker(f"""This is a {tokenizer.mask_token}""" )
self.assertEqual(
_UpperCAmelCase , [
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
] , )
UpperCamelCase_ = {vocab[el] for el in targets}
self.assertEqual({el['token'] for el in outputs} , _UpperCAmelCase )
UpperCamelCase_ = [tokenizer.decode([x] ) for x in target_ids]
self.assertEqual({el['token_str'] for el in outputs} , set(_UpperCAmelCase ) )
# Call argument
UpperCamelCase_ = FillMaskPipeline(model=_UpperCAmelCase , tokenizer=_UpperCAmelCase )
UpperCamelCase_ = fill_masker(f"""This is a {tokenizer.mask_token}""" , targets=_UpperCAmelCase )
self.assertEqual(
_UpperCAmelCase , [
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
] , )
UpperCamelCase_ = {vocab[el] for el in targets}
self.assertEqual({el['token'] for el in outputs} , _UpperCAmelCase )
UpperCamelCase_ = [tokenizer.decode([x] ) for x in target_ids]
self.assertEqual({el['token_str'] for el in outputs} , set(_UpperCAmelCase ) )
# Score equivalence
UpperCamelCase_ = fill_masker(f"""This is a {tokenizer.mask_token}""" , targets=_UpperCAmelCase )
UpperCamelCase_ = [top_mask['token_str'] for top_mask in outputs]
UpperCamelCase_ = [top_mask['score'] for top_mask in outputs]
# For some BPE tokenizers, `</w>` is removed during decoding, so `token_str` won't be the same as in `targets`.
if set(_UpperCAmelCase ) == set(_UpperCAmelCase ):
UpperCamelCase_ = fill_masker(f"""This is a {tokenizer.mask_token}""" , targets=_UpperCAmelCase )
UpperCamelCase_ = [top_mask['score'] for top_mask in unmasked_targets]
self.assertEqual(nested_simplify(_UpperCAmelCase ) , nested_simplify(_UpperCAmelCase ) )
# Raises with invalid
with self.assertRaises(_UpperCAmelCase ):
UpperCamelCase_ = fill_masker(f"""This is a {tokenizer.mask_token}""" , targets=[] )
# For some tokenizers, `""` is actually in the vocabulary and the expected error won't raised
if "" not in tokenizer.get_vocab():
with self.assertRaises(_UpperCAmelCase ):
UpperCamelCase_ = fill_masker(f"""This is a {tokenizer.mask_token}""" , targets=[''] )
with self.assertRaises(_UpperCAmelCase ):
UpperCamelCase_ = fill_masker(f"""This is a {tokenizer.mask_token}""" , targets='' )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> Tuple:
UpperCamelCase_ = FillMaskPipeline(model=_UpperCAmelCase , tokenizer=_UpperCAmelCase , top_k=2 )
UpperCamelCase_ = fill_masker(f"""This is a {tokenizer.mask_token}""" )
self.assertEqual(
_UpperCAmelCase , [
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
] , )
UpperCamelCase_ = FillMaskPipeline(model=_UpperCAmelCase , tokenizer=_UpperCAmelCase )
UpperCamelCase_ = fill_masker(f"""This is a {tokenizer.mask_token}""" , top_k=2 )
self.assertEqual(
_UpperCAmelCase , [
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
] , )
self.assertEqual(nested_simplify(_UpperCAmelCase ) , nested_simplify(_UpperCAmelCase ) )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> List[Any]:
UpperCamelCase_ = tokenizer.get_vocab()
UpperCamelCase_ = FillMaskPipeline(model=_UpperCAmelCase , tokenizer=_UpperCAmelCase )
# top_k=2, ntargets=3
UpperCamelCase_ = sorted(vocab.keys() )[:3]
UpperCamelCase_ = fill_masker(f"""This is a {tokenizer.mask_token}""" , top_k=2 , targets=_UpperCAmelCase )
# If we use the most probably targets, and filter differently, we should still
# have the same results
UpperCamelCase_ = [el['token_str'] for el in sorted(_UpperCAmelCase , key=lambda _UpperCAmelCase : x["score"] , reverse=_UpperCAmelCase )]
# For some BPE tokenizers, `</w>` is removed during decoding, so `token_str` won't be the same as in `targets`.
if set(_UpperCAmelCase ).issubset(_UpperCAmelCase ):
UpperCamelCase_ = fill_masker(f"""This is a {tokenizer.mask_token}""" , top_k=3 , targets=_UpperCAmelCase )
# They should yield exactly the same result
self.assertEqual(nested_simplify(_UpperCAmelCase ) , nested_simplify(_UpperCAmelCase ) )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> Optional[int]:
UpperCamelCase_ = FillMaskPipeline(model=_UpperCAmelCase , tokenizer=_UpperCAmelCase )
UpperCamelCase_ = tokenizer.get_vocab()
# String duplicates + id duplicates
UpperCamelCase_ = sorted(vocab.keys() )[:3]
UpperCamelCase_ = [targets[0], targets[1], targets[0], targets[2], targets[1]]
UpperCamelCase_ = fill_masker(f"""My name is {tokenizer.mask_token}""" , targets=_UpperCAmelCase , top_k=10 )
# The target list contains duplicates, so we can't output more
# than them
self.assertEqual(len(_UpperCAmelCase ) , 3 )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> List[str]:
UpperCamelCase_ = FillMaskPipeline(model=_UpperCAmelCase , tokenizer=_UpperCAmelCase )
UpperCamelCase_ = fill_masker(
f"""This is a {tokenizer.mask_token} {tokenizer.mask_token} {tokenizer.mask_token}""" , top_k=2 )
self.assertEqual(
_UpperCAmelCase , [
[
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
],
[
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
],
[
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
],
] , )
| 23 | 0 |
from dataclasses import dataclass
from typing import List, Optional, Union
import numpy as np
import PIL
import torch
from transformers import CLIPImageProcessor, CLIPVisionModel
from ...models import PriorTransformer
from ...pipelines import DiffusionPipeline
from ...schedulers import HeunDiscreteScheduler
from ...utils import (
BaseOutput,
is_accelerate_available,
logging,
randn_tensor,
replace_example_docstring,
)
from .renderer import ShapERenderer
UpperCAmelCase_ = logging.get_logger(__name__) # pylint: disable=invalid-name
UpperCAmelCase_ = """
Examples:
```py
>>> from PIL import Image
>>> import torch
>>> from diffusers import DiffusionPipeline
>>> from diffusers.utils import export_to_gif, load_image
>>> device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")
>>> repo = \"openai/shap-e-img2img\"
>>> pipe = DiffusionPipeline.from_pretrained(repo, torch_dtype=torch.float16)
>>> pipe = pipe.to(device)
>>> guidance_scale = 3.0
>>> image_url = \"https://hf.co/datasets/diffusers/docs-images/resolve/main/shap-e/corgi.png\"
>>> image = load_image(image_url).convert(\"RGB\")
>>> images = pipe(
... image,
... guidance_scale=guidance_scale,
... num_inference_steps=64,
... frame_size=256,
... ).images
>>> gif_path = export_to_gif(images[0], \"corgi_3d.gif\")
```
"""
@dataclass
class lowerCamelCase__ ( _A):
"""simple docstring"""
a__ : Union[PIL.Image.Image, np.ndarray]
class lowerCamelCase__ ( _A):
"""simple docstring"""
def __init__( self : int , __lowerCAmelCase : PriorTransformer , __lowerCAmelCase : CLIPVisionModel , __lowerCAmelCase : CLIPImageProcessor , __lowerCAmelCase : HeunDiscreteScheduler , __lowerCAmelCase : ShapERenderer , ) -> List[str]:
super().__init__()
self.register_modules(
prior=__lowerCAmelCase , image_encoder=__lowerCAmelCase , image_processor=__lowerCAmelCase , scheduler=__lowerCAmelCase , renderer=__lowerCAmelCase , )
def snake_case_ ( self : Optional[int] , __lowerCAmelCase : Optional[Any] , __lowerCAmelCase : List[Any] , __lowerCAmelCase : Optional[int] , __lowerCAmelCase : List[str] , __lowerCAmelCase : str , __lowerCAmelCase : Tuple ) -> int:
if latents is None:
_A = randn_tensor(__lowerCAmelCase , generator=__lowerCAmelCase , device=__lowerCAmelCase , dtype=__lowerCAmelCase )
else:
if latents.shape != shape:
raise ValueError(f'''Unexpected latents shape, got {latents.shape}, expected {shape}''' )
_A = latents.to(__lowerCAmelCase )
_A = latents * scheduler.init_noise_sigma
return latents
def snake_case_ ( self : str , __lowerCAmelCase : List[Any]=0 ) -> Optional[Any]:
if is_accelerate_available():
from accelerate import cpu_offload
else:
raise ImportError('''Please install accelerate via `pip install accelerate`''' )
_A = torch.device(f'''cuda:{gpu_id}''' )
_A = [self.image_encoder, self.prior]
for cpu_offloaded_model in models:
if cpu_offloaded_model is not None:
cpu_offload(__lowerCAmelCase , __lowerCAmelCase )
@property
def snake_case_ ( self : int ) -> List[Any]:
if self.device != torch.device('''meta''' ) or not hasattr(self.image_encoder , '''_hf_hook''' ):
return self.device
for module in self.image_encoder.modules():
if (
hasattr(__lowerCAmelCase , '''_hf_hook''' )
and hasattr(module._hf_hook , '''execution_device''' )
and module._hf_hook.execution_device is not None
):
return torch.device(module._hf_hook.execution_device )
return self.device
def snake_case_ ( self : List[str] , __lowerCAmelCase : str , __lowerCAmelCase : Optional[Any] , __lowerCAmelCase : List[Any] , __lowerCAmelCase : Dict , ) -> Optional[int]:
if isinstance(__lowerCAmelCase , __lowerCAmelCase ) and isinstance(image[0] , torch.Tensor ):
_A = torch.cat(__lowerCAmelCase , axis=0 ) if image[0].ndim == 4 else torch.stack(__lowerCAmelCase , axis=0 )
if not isinstance(__lowerCAmelCase , torch.Tensor ):
_A = self.image_processor(__lowerCAmelCase , return_tensors='''pt''' ).pixel_values[0].unsqueeze(0 )
_A = image.to(dtype=self.image_encoder.dtype , device=__lowerCAmelCase )
_A = self.image_encoder(__lowerCAmelCase )['''last_hidden_state''']
_A = image_embeds[:, 1:, :].contiguous() # batch_size, dim, 256
_A = image_embeds.repeat_interleave(__lowerCAmelCase , dim=0 )
if do_classifier_free_guidance:
_A = torch.zeros_like(__lowerCAmelCase )
# For classifier free guidance, we need to do two forward passes.
# Here we concatenate the unconditional and text embeddings into a single batch
# to avoid doing two forward passes
_A = torch.cat([negative_image_embeds, image_embeds] )
return image_embeds
@torch.no_grad()
@replace_example_docstring(__lowerCAmelCase )
def __call__( self : str , __lowerCAmelCase : Union[PIL.Image.Image, List[PIL.Image.Image]] , __lowerCAmelCase : int = 1 , __lowerCAmelCase : int = 25 , __lowerCAmelCase : Optional[Union[torch.Generator, List[torch.Generator]]] = None , __lowerCAmelCase : Optional[torch.FloatTensor] = None , __lowerCAmelCase : float = 4.0 , __lowerCAmelCase : int = 64 , __lowerCAmelCase : Optional[str] = "pil" , __lowerCAmelCase : bool = True , ) -> Tuple:
if isinstance(__lowerCAmelCase , PIL.Image.Image ):
_A = 1
elif isinstance(__lowerCAmelCase , torch.Tensor ):
_A = image.shape[0]
elif isinstance(__lowerCAmelCase , __lowerCAmelCase ) and isinstance(image[0] , (torch.Tensor, PIL.Image.Image) ):
_A = len(__lowerCAmelCase )
else:
raise ValueError(
f'''`image` has to be of type `PIL.Image.Image`, `torch.Tensor`, `List[PIL.Image.Image]` or `List[torch.Tensor]` but is {type(__lowerCAmelCase )}''' )
_A = self._execution_device
_A = batch_size * num_images_per_prompt
_A = guidance_scale > 1.0
_A = self._encode_image(__lowerCAmelCase , __lowerCAmelCase , __lowerCAmelCase , __lowerCAmelCase )
# prior
self.scheduler.set_timesteps(__lowerCAmelCase , device=__lowerCAmelCase )
_A = self.scheduler.timesteps
_A = self.prior.config.num_embeddings
_A = self.prior.config.embedding_dim
_A = self.prepare_latents(
(batch_size, num_embeddings * embedding_dim) , image_embeds.dtype , __lowerCAmelCase , __lowerCAmelCase , __lowerCAmelCase , self.scheduler , )
# YiYi notes: for testing only to match ldm, we can directly create a latents with desired shape: batch_size, num_embeddings, embedding_dim
_A = latents.reshape(latents.shape[0] , __lowerCAmelCase , __lowerCAmelCase )
for i, t in enumerate(self.progress_bar(__lowerCAmelCase ) ):
# expand the latents if we are doing classifier free guidance
_A = torch.cat([latents] * 2 ) if do_classifier_free_guidance else latents
_A = self.scheduler.scale_model_input(__lowerCAmelCase , __lowerCAmelCase )
_A = self.prior(
__lowerCAmelCase , timestep=__lowerCAmelCase , proj_embedding=__lowerCAmelCase , ).predicted_image_embedding
# remove the variance
_A , _A = noise_pred.split(
scaled_model_input.shape[2] , dim=2 ) # batch_size, num_embeddings, embedding_dim
if do_classifier_free_guidance is not None:
_A , _A = noise_pred.chunk(2 )
_A = noise_pred_uncond + guidance_scale * (noise_pred - noise_pred_uncond)
_A = self.scheduler.step(
__lowerCAmelCase , timestep=__lowerCAmelCase , sample=__lowerCAmelCase , ).prev_sample
if output_type == "latent":
return ShapEPipelineOutput(images=__lowerCAmelCase )
_A = []
for i, latent in enumerate(__lowerCAmelCase ):
print()
_A = self.renderer.decode(
latent[None, :] , __lowerCAmelCase , size=__lowerCAmelCase , ray_batch_size=40_96 , n_coarse_samples=64 , n_fine_samples=1_28 , )
images.append(__lowerCAmelCase )
_A = torch.stack(__lowerCAmelCase )
if output_type not in ["np", "pil"]:
raise ValueError(f'''Only the output types `pil` and `np` are supported not output_type={output_type}''' )
_A = images.cpu().numpy()
if output_type == "pil":
_A = [self.numpy_to_pil(__lowerCAmelCase ) for image in images]
# Offload last model to CPU
if hasattr(self , '''final_offload_hook''' ) and self.final_offload_hook is not None:
self.final_offload_hook.offload()
if not return_dict:
return (images,)
return ShapEPipelineOutput(images=__lowerCAmelCase )
| 2 |
import gc
import unittest
import numpy as np
import torch
from transformers import CLIPTextConfig, CLIPTextModel, CLIPTokenizer
from diffusers import (
AutoencoderKL,
DDIMScheduler,
StableDiffusionSAGPipeline,
UNetaDConditionModel,
)
from diffusers.utils import slow, torch_device
from diffusers.utils.testing_utils import enable_full_determinism, require_torch_gpu
from ..pipeline_params import TEXT_TO_IMAGE_BATCH_PARAMS, TEXT_TO_IMAGE_IMAGE_PARAMS, TEXT_TO_IMAGE_PARAMS
from ..test_pipelines_common import PipelineLatentTesterMixin, PipelineTesterMixin
enable_full_determinism()
class _a ( UpperCAmelCase__ , UpperCAmelCase__ , unittest.TestCase ):
"""simple docstring"""
A_ = StableDiffusionSAGPipeline
A_ = TEXT_TO_IMAGE_PARAMS
A_ = TEXT_TO_IMAGE_BATCH_PARAMS
A_ = TEXT_TO_IMAGE_IMAGE_PARAMS
A_ = TEXT_TO_IMAGE_IMAGE_PARAMS
A_ = False
def _UpperCAmelCase ( self ) -> Optional[Any]:
torch.manual_seed(0 )
UpperCamelCase_ = UNetaDConditionModel(
block_out_channels=(32, 64) , layers_per_block=2 , sample_size=32 , in_channels=4 , out_channels=4 , down_block_types=('DownBlock2D', 'CrossAttnDownBlock2D') , up_block_types=('CrossAttnUpBlock2D', 'UpBlock2D') , cross_attention_dim=32 , )
UpperCamelCase_ = DDIMScheduler(
beta_start=0.0_0_0_8_5 , beta_end=0.0_1_2 , beta_schedule='scaled_linear' , clip_sample=_UpperCAmelCase , set_alpha_to_one=_UpperCAmelCase , )
torch.manual_seed(0 )
UpperCamelCase_ = AutoencoderKL(
block_out_channels=[32, 64] , in_channels=3 , out_channels=3 , down_block_types=['DownEncoderBlock2D', 'DownEncoderBlock2D'] , up_block_types=['UpDecoderBlock2D', 'UpDecoderBlock2D'] , latent_channels=4 , )
torch.manual_seed(0 )
UpperCamelCase_ = CLIPTextConfig(
bos_token_id=0 , eos_token_id=2 , hidden_size=32 , intermediate_size=37 , layer_norm_eps=1e-05 , num_attention_heads=4 , num_hidden_layers=5 , pad_token_id=1 , vocab_size=1000 , )
UpperCamelCase_ = CLIPTextModel(_UpperCAmelCase )
UpperCamelCase_ = CLIPTokenizer.from_pretrained('hf-internal-testing/tiny-random-clip' )
UpperCamelCase_ = {
'unet': unet,
'scheduler': scheduler,
'vae': vae,
'text_encoder': text_encoder,
'tokenizer': tokenizer,
'safety_checker': None,
'feature_extractor': None,
}
return components
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase=0 ) -> List[Any]:
if str(_UpperCAmelCase ).startswith('mps' ):
UpperCamelCase_ = torch.manual_seed(_UpperCAmelCase )
else:
UpperCamelCase_ = torch.Generator(device=_UpperCAmelCase ).manual_seed(_UpperCAmelCase )
UpperCamelCase_ = {
'prompt': '.',
'generator': generator,
'num_inference_steps': 2,
'guidance_scale': 1.0,
'sag_scale': 1.0,
'output_type': 'numpy',
}
return inputs
def _UpperCAmelCase ( self ) -> Tuple:
super().test_inference_batch_single_identical(expected_max_diff=3e-3 )
@slow
@require_torch_gpu
class _a ( unittest.TestCase ):
"""simple docstring"""
def _UpperCAmelCase ( self ) -> Tuple:
# clean up the VRAM after each test
super().tearDown()
gc.collect()
torch.cuda.empty_cache()
def _UpperCAmelCase ( self ) -> str:
UpperCamelCase_ = StableDiffusionSAGPipeline.from_pretrained('CompVis/stable-diffusion-v1-4' )
UpperCamelCase_ = sag_pipe.to(_UpperCAmelCase )
sag_pipe.set_progress_bar_config(disable=_UpperCAmelCase )
UpperCamelCase_ = '.'
UpperCamelCase_ = torch.manual_seed(0 )
UpperCamelCase_ = sag_pipe(
[prompt] , generator=_UpperCAmelCase , guidance_scale=7.5 , sag_scale=1.0 , num_inference_steps=20 , output_type='np' )
UpperCamelCase_ = output.images
UpperCamelCase_ = image[0, -3:, -3:, -1]
assert image.shape == (1, 512, 512, 3)
UpperCamelCase_ = np.array([0.1_5_6_8, 0.1_7_3_8, 0.1_6_9_5, 0.1_6_9_3, 0.1_5_0_7, 0.1_7_0_5, 0.1_5_4_7, 0.1_7_5_1, 0.1_9_4_9] )
assert np.abs(image_slice.flatten() - expected_slice ).max() < 5e-2
def _UpperCAmelCase ( self ) -> List[Any]:
UpperCamelCase_ = StableDiffusionSAGPipeline.from_pretrained('stabilityai/stable-diffusion-2-1-base' )
UpperCamelCase_ = sag_pipe.to(_UpperCAmelCase )
sag_pipe.set_progress_bar_config(disable=_UpperCAmelCase )
UpperCamelCase_ = '.'
UpperCamelCase_ = torch.manual_seed(0 )
UpperCamelCase_ = sag_pipe(
[prompt] , generator=_UpperCAmelCase , guidance_scale=7.5 , sag_scale=1.0 , num_inference_steps=20 , output_type='np' )
UpperCamelCase_ = output.images
UpperCamelCase_ = image[0, -3:, -3:, -1]
assert image.shape == (1, 512, 512, 3)
UpperCamelCase_ = np.array([0.3_4_5_9, 0.2_8_7_6, 0.2_5_3_7, 0.3_0_0_2, 0.2_6_7_1, 0.2_1_6_0, 0.3_0_2_6, 0.2_2_6_2, 0.2_3_7_1] )
assert np.abs(image_slice.flatten() - expected_slice ).max() < 5e-2
def _UpperCAmelCase ( self ) -> Dict:
UpperCamelCase_ = StableDiffusionSAGPipeline.from_pretrained('stabilityai/stable-diffusion-2-1-base' )
UpperCamelCase_ = sag_pipe.to(_UpperCAmelCase )
sag_pipe.set_progress_bar_config(disable=_UpperCAmelCase )
UpperCamelCase_ = '.'
UpperCamelCase_ = torch.manual_seed(0 )
UpperCamelCase_ = sag_pipe(
[prompt] , width=768 , height=512 , generator=_UpperCAmelCase , guidance_scale=7.5 , sag_scale=1.0 , num_inference_steps=20 , output_type='np' , )
UpperCamelCase_ = output.images
assert image.shape == (1, 512, 768, 3)
| 23 | 0 |
'''simple docstring'''
import argparse
import os
from pathlib import Path
from typing import Dict
import tensorflow as tf
import torch
from tqdm import tqdm
from transformers import PegasusConfig, PegasusForConditionalGeneration, PegasusTokenizer
from transformers.models.pegasus.configuration_pegasus import DEFAULTS, task_specific_params
lowerCAmelCase : Union[str, Any] = [
# replace left string with right string to get the relevant state_dict key (identical state dict to bart)
['memory_attention', 'encoder_attn'],
['attention', 'attn'],
['/', '.'],
['.LayerNorm.gamma', '_layer_norm.weight'],
['.LayerNorm.beta', '_layer_norm.bias'],
['r.layer_', 'r.layers.'],
['output_proj', 'out_proj'],
['ffn.dense_1.', 'fc2.'],
['ffn.dense.', 'fc1.'],
['ffn_layer_norm', 'final_layer_norm'],
['kernel', 'weight'],
['encoder_layer_norm.', 'encoder.layer_norm.'],
['decoder_layer_norm.', 'decoder.layer_norm.'],
['embeddings.weights', 'shared.weight'],
]
def A_( A : Dict):
for pegasus_name, hf_name in PATTERNS:
UpperCamelCase = k.replace(A , A)
return k
def A_( A : dict , A : dict):
UpperCamelCase = DEFAULTS.copy()
cfg_kwargs.update(A)
UpperCamelCase = PegasusConfig(**A)
UpperCamelCase = PegasusForConditionalGeneration(A)
UpperCamelCase = torch_model.model.state_dict()
UpperCamelCase = {}
for k, v in tf_weights.items():
UpperCamelCase = rename_state_dict_key(A)
if new_k not in sd:
raise ValueError(f'''could not find new key {new_k} in state dict. (converted from {k})''')
if "dense" in k or "proj" in new_k:
UpperCamelCase = v.T
UpperCamelCase = torch.tensor(A , dtype=sd[new_k].dtype)
assert v.shape == sd[new_k].shape, f'''{new_k}, {k}, {v.shape}, {sd[new_k].shape}'''
# make sure embedding.padding_idx is respected
UpperCamelCase = torch.zeros_like(mapping['shared.weight'][cfg.pad_token_id + 1])
UpperCamelCase = mapping['shared.weight']
UpperCamelCase = mapping['shared.weight']
UpperCamelCase = {k: torch.zeros_like(A) for k, v in sd.items() if k.endswith('bias') and k not in mapping}
mapping.update(**A)
UpperCamelCase , UpperCamelCase = torch_model.model.load_state_dict(A , strict=A)
UpperCamelCase = [
k for k in missing if k not in ['encoder.embed_positions.weight', 'decoder.embed_positions.weight']
]
assert unexpected_missing == [], f'''no matches found for the following torch keys {unexpected_missing}'''
assert extra == [], f'''no matches found for the following tf keys {extra}'''
return torch_model
def A_( A : Tuple="./ckpt/aeslc/model.ckpt-32000"):
UpperCamelCase = tf.train.list_variables(A)
UpperCamelCase = {}
UpperCamelCase = ['Adafactor', 'global_step']
for name, shape in tqdm(A , desc='converting tf checkpoint to dict'):
UpperCamelCase = any(pat in name for pat in ignore_name)
if skip_key:
continue
UpperCamelCase = tf.train.load_variable(A , A)
UpperCamelCase = array
return tf_weights
def A_( A : str , A : str):
# save tokenizer first
UpperCamelCase = Path(A).parent.name
UpperCamelCase = task_specific_params[f'''summarization_{dataset}''']['max_position_embeddings']
UpperCamelCase = PegasusTokenizer.from_pretrained('sshleifer/pegasus' , model_max_length=A)
assert tok.model_max_length == desired_max_model_length
tok.save_pretrained(A)
# convert model
UpperCamelCase = get_tf_weights_as_numpy(A)
UpperCamelCase = task_specific_params[f'''summarization_{dataset}''']
if dataset == "large":
UpperCamelCase = task_specific_params
UpperCamelCase = convert_pegasus(A , A)
torch_model.save_pretrained(A)
UpperCamelCase = torch_model.state_dict()
sd.pop('model.decoder.embed_positions.weight')
sd.pop('model.encoder.embed_positions.weight')
torch.save(A , Path(A) / 'pytorch_model.bin')
if __name__ == "__main__":
lowerCAmelCase : List[str] = argparse.ArgumentParser()
# Required parameters
parser.add_argument('tf_ckpt_path', type=str, help='passed to tf.train.list_variables')
parser.add_argument('save_dir', default=None, type=str, help='Path to the output PyTorch model.')
lowerCAmelCase : Any = parser.parse_args()
if args.save_dir is None:
lowerCAmelCase : Optional[int] = Path(args.tf_ckpt_path).parent.name
lowerCAmelCase : List[Any] = os.path.join('pegasus', dataset)
convert_pegasus_ckpt_to_pytorch(args.tf_ckpt_path, args.save_dir)
| 3 |
from __future__ import annotations
from sys import maxsize
from typing import Generic, TypeVar
snake_case__ : List[str] = TypeVar("""T""")
def _snake_case (__lowercase):
return (position - 1) // 2
def _snake_case (__lowercase):
return (2 * position) + 1
def _snake_case (__lowercase):
return (2 * position) + 2
class _a ( Generic[T] ):
"""simple docstring"""
def __init__( self ) -> None:
UpperCamelCase_ = []
UpperCamelCase_ = {}
UpperCamelCase_ = 0
def __len__( self ) -> int:
return self.elements
def __repr__( self ) -> str:
return str(self.heap )
def _UpperCAmelCase ( self ) -> bool:
# Check if the priority queue is empty
return self.elements == 0
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> None:
# Add an element with given priority to the queue
self.heap.append((elem, weight) )
UpperCamelCase_ = self.elements
self.elements += 1
self._bubble_up(_UpperCAmelCase )
def _UpperCAmelCase ( self ) -> T:
# Remove and return the element with lowest weight (highest priority)
if self.elements > 1:
self._swap_nodes(0 , self.elements - 1 )
UpperCamelCase_ , UpperCamelCase_ = self.heap.pop()
del self.position_map[elem]
self.elements -= 1
if self.elements > 0:
UpperCamelCase_ , UpperCamelCase_ = self.heap[0]
self._bubble_down(_UpperCAmelCase )
return elem
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> None:
# Update the weight of the given key
UpperCamelCase_ = self.position_map[elem]
UpperCamelCase_ = (elem, weight)
if position > 0:
UpperCamelCase_ = get_parent_position(_UpperCAmelCase )
UpperCamelCase_ , UpperCamelCase_ = self.heap[parent_position]
if parent_weight > weight:
self._bubble_up(_UpperCAmelCase )
else:
self._bubble_down(_UpperCAmelCase )
else:
self._bubble_down(_UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase ) -> None:
# Place a node at the proper position (upward movement) [to be used internally
# only]
UpperCamelCase_ = self.position_map[elem]
if curr_pos == 0:
return None
UpperCamelCase_ = get_parent_position(_UpperCAmelCase )
UpperCamelCase_ , UpperCamelCase_ = self.heap[curr_pos]
UpperCamelCase_ , UpperCamelCase_ = self.heap[parent_position]
if parent_weight > weight:
self._swap_nodes(_UpperCAmelCase , _UpperCAmelCase )
return self._bubble_up(_UpperCAmelCase )
return None
def _UpperCAmelCase ( self , _UpperCAmelCase ) -> None:
# Place a node at the proper position (downward movement) [to be used
# internally only]
UpperCamelCase_ = self.position_map[elem]
UpperCamelCase_ , UpperCamelCase_ = self.heap[curr_pos]
UpperCamelCase_ = get_child_left_position(_UpperCAmelCase )
UpperCamelCase_ = get_child_right_position(_UpperCAmelCase )
if child_left_position < self.elements and child_right_position < self.elements:
UpperCamelCase_ , UpperCamelCase_ = self.heap[child_left_position]
UpperCamelCase_ , UpperCamelCase_ = self.heap[child_right_position]
if child_right_weight < child_left_weight and child_right_weight < weight:
self._swap_nodes(_UpperCAmelCase , _UpperCAmelCase )
return self._bubble_down(_UpperCAmelCase )
if child_left_position < self.elements:
UpperCamelCase_ , UpperCamelCase_ = self.heap[child_left_position]
if child_left_weight < weight:
self._swap_nodes(_UpperCAmelCase , _UpperCAmelCase )
return self._bubble_down(_UpperCAmelCase )
else:
return None
if child_right_position < self.elements:
UpperCamelCase_ , UpperCamelCase_ = self.heap[child_right_position]
if child_right_weight < weight:
self._swap_nodes(_UpperCAmelCase , _UpperCAmelCase )
return self._bubble_down(_UpperCAmelCase )
return None
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> None:
# Swap the nodes at the given positions
UpperCamelCase_ = self.heap[nodea_pos][0]
UpperCamelCase_ = self.heap[nodea_pos][0]
UpperCamelCase_ , UpperCamelCase_ = (
self.heap[nodea_pos],
self.heap[nodea_pos],
)
UpperCamelCase_ = nodea_pos
UpperCamelCase_ = nodea_pos
class _a ( Generic[T] ):
"""simple docstring"""
def __init__( self ) -> None:
UpperCamelCase_ = {}
UpperCamelCase_ = 0
def __repr__( self ) -> str:
return str(self.connections )
def __len__( self ) -> int:
return self.nodes
def _UpperCAmelCase ( self , _UpperCAmelCase ) -> None:
# Add a node in the graph if it is not in the graph
if node not in self.connections:
UpperCamelCase_ = {}
self.nodes += 1
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase ) -> None:
# Add an edge between 2 nodes in the graph
self.add_node(_UpperCAmelCase )
self.add_node(_UpperCAmelCase )
UpperCamelCase_ = weight
UpperCamelCase_ = weight
def _snake_case (__lowercase , ):
UpperCamelCase_ = {node: maxsize for node in graph.connections}
UpperCamelCase_ = {node: None for node in graph.connections}
UpperCamelCase_ = MinPriorityQueue()
for node, weight in dist.items():
priority_queue.push(__lowercase , __lowercase)
if priority_queue.is_empty():
return dist, parent
# initialization
UpperCamelCase_ = priority_queue.extract_min()
UpperCamelCase_ = 0
for neighbour in graph.connections[node]:
if dist[neighbour] > dist[node] + graph.connections[node][neighbour]:
UpperCamelCase_ = dist[node] + graph.connections[node][neighbour]
priority_queue.update_key(__lowercase , dist[neighbour])
UpperCamelCase_ = node
# running prim's algorithm
while not priority_queue.is_empty():
UpperCamelCase_ = priority_queue.extract_min()
for neighbour in graph.connections[node]:
if dist[neighbour] > dist[node] + graph.connections[node][neighbour]:
UpperCamelCase_ = dist[node] + graph.connections[node][neighbour]
priority_queue.update_key(__lowercase , dist[neighbour])
UpperCamelCase_ = node
return dist, parent
| 23 | 0 |
"""simple docstring"""
class a :
def __init__( self , _snake_case ):
"""simple docstring"""
lowerCAmelCase = size
lowerCAmelCase = [0] * size
lowerCAmelCase = [0] * size
@staticmethod
def UpperCamelCase__ ( _snake_case ):
"""simple docstring"""
return index | (index + 1)
@staticmethod
def UpperCamelCase__ ( _snake_case ):
"""simple docstring"""
return (index & (index + 1)) - 1
def UpperCamelCase__ ( self , _snake_case , _snake_case ):
"""simple docstring"""
lowerCAmelCase = value
while index < self.size:
lowerCAmelCase = self.get_prev(_snake_case ) + 1
if current_left_border == index:
lowerCAmelCase = value
else:
lowerCAmelCase = max(_snake_case , _snake_case , _snake_case )
lowerCAmelCase = self.get_next(_snake_case )
def UpperCamelCase__ ( self , _snake_case , _snake_case ):
"""simple docstring"""
right -= 1 # Because of right is exclusive
lowerCAmelCase = 0
while left <= right:
lowerCAmelCase = self.get_prev(_snake_case )
if left <= current_left:
lowerCAmelCase = max(_snake_case , self.tree[right] )
lowerCAmelCase = current_left
else:
lowerCAmelCase = max(_snake_case , self.arr[right] )
right -= 1
return result
if __name__ == "__main__":
import doctest
doctest.testmod()
| 4 |
from __future__ import annotations
import sys
from collections import deque
from typing import Generic, TypeVar
snake_case__ : Dict = TypeVar("""T""")
class _a ( Generic[T] ):
"""simple docstring"""
A_ = 42 # Cache store of keys
A_ = 42 # References of the keys in cache
A_ = 10 # Maximum capacity of cache
def __init__( self , _UpperCAmelCase ) -> None:
UpperCamelCase_ = deque()
UpperCamelCase_ = set()
if not n:
UpperCamelCase_ = sys.maxsize
elif n < 0:
raise ValueError('n should be an integer greater than 0.' )
else:
UpperCamelCase_ = n
def _UpperCAmelCase ( self , _UpperCAmelCase ) -> None:
if x not in self.key_reference:
if len(self.dq_store ) == LRUCache._MAX_CAPACITY:
UpperCamelCase_ = self.dq_store.pop()
self.key_reference.remove(_UpperCAmelCase )
else:
self.dq_store.remove(_UpperCAmelCase )
self.dq_store.appendleft(_UpperCAmelCase )
self.key_reference.add(_UpperCAmelCase )
def _UpperCAmelCase ( self ) -> None:
for k in self.dq_store:
print(_UpperCAmelCase )
def __repr__( self ) -> str:
return f"""LRUCache({self._MAX_CAPACITY}) => {list(self.dq_store )}"""
if __name__ == "__main__":
import doctest
doctest.testmod()
snake_case__ : LRUCache[str | int] = LRUCache(4)
lru_cache.refer("""A""")
lru_cache.refer(2)
lru_cache.refer(3)
lru_cache.refer("""A""")
lru_cache.refer(4)
lru_cache.refer(5)
lru_cache.display()
print(lru_cache)
assert str(lru_cache) == "LRUCache(4) => [5, 4, 'A', 3]"
| 23 | 0 |
'''simple docstring'''
from dataclasses import dataclass
from typing import Optional, Tuple, Union
import numpy as np
import torch
from ..configuration_utils import ConfigMixin, register_to_config
from ..utils import BaseOutput, randn_tensor
from .scheduling_utils import SchedulerMixin
@dataclass
class UpperCAmelCase_ ( _SCREAMING_SNAKE_CASE ):
'''simple docstring'''
_lowercase : torch.FloatTensor
_lowercase : torch.FloatTensor
_lowercase : Optional[torch.FloatTensor] = None
class UpperCAmelCase_ ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ):
'''simple docstring'''
_lowercase : Optional[int] = 2
@register_to_config
def __init__( self , _lowercase = 0.02 , _lowercase = 100 , _lowercase = 1.007 , _lowercase = 80 , _lowercase = 0.05 , _lowercase = 50 , ):
"""simple docstring"""
_lowerCAmelCase = sigma_max
# setable values
_lowerCAmelCase = None
_lowerCAmelCase = None
_lowerCAmelCase = None # sigma(t_i)
def _lowercase ( self , _lowercase , _lowercase = None ):
"""simple docstring"""
return sample
def _lowercase ( self , _lowercase , _lowercase = None ):
"""simple docstring"""
_lowerCAmelCase = num_inference_steps
_lowerCAmelCase = np.arange(0 , self.num_inference_steps )[::-1].copy()
_lowerCAmelCase = torch.from_numpy(_lowercase ).to(_lowercase )
_lowerCAmelCase = [
(
self.config.sigma_max**2
* (self.config.sigma_min**2 / self.config.sigma_max**2) ** (i / (num_inference_steps - 1))
)
for i in self.timesteps
]
_lowerCAmelCase = torch.tensor(_lowercase , dtype=torch.floataa , device=_lowercase )
def _lowercase ( self , _lowercase , _lowercase , _lowercase = None ):
"""simple docstring"""
if self.config.s_min <= sigma <= self.config.s_max:
_lowerCAmelCase = min(self.config.s_churn / self.num_inference_steps , 2**0.5 - 1 )
else:
_lowerCAmelCase = 0
# sample eps ~ N(0, S_noise^2 * I)
_lowerCAmelCase = self.config.s_noise * randn_tensor(sample.shape , generator=_lowercase ).to(sample.device )
_lowerCAmelCase = sigma + gamma * sigma
_lowerCAmelCase = sample + ((sigma_hat**2 - sigma**2) ** 0.5 * eps)
return sample_hat, sigma_hat
def _lowercase ( self , _lowercase , _lowercase , _lowercase , _lowercase , _lowercase = True , ):
"""simple docstring"""
_lowerCAmelCase = sample_hat + sigma_hat * model_output
_lowerCAmelCase = (sample_hat - pred_original_sample) / sigma_hat
_lowerCAmelCase = sample_hat + (sigma_prev - sigma_hat) * derivative
if not return_dict:
return (sample_prev, derivative)
return KarrasVeOutput(
prev_sample=_lowercase , derivative=_lowercase , pred_original_sample=_lowercase )
def _lowercase ( self , _lowercase , _lowercase , _lowercase , _lowercase , _lowercase , _lowercase , _lowercase = True , ):
"""simple docstring"""
_lowerCAmelCase = sample_prev + sigma_prev * model_output
_lowerCAmelCase = (sample_prev - pred_original_sample) / sigma_prev
_lowerCAmelCase = sample_hat + (sigma_prev - sigma_hat) * (0.5 * derivative + 0.5 * derivative_corr)
if not return_dict:
return (sample_prev, derivative)
return KarrasVeOutput(
prev_sample=_lowercase , derivative=_lowercase , pred_original_sample=_lowercase )
def _lowercase ( self , _lowercase , _lowercase , _lowercase ):
"""simple docstring"""
raise NotImplementedError()
| 5 |
import numpy as np
def _snake_case (__lowercase):
return 1 / (1 + np.exp(-vector))
def _snake_case (__lowercase):
return vector * sigmoid(__lowercase)
if __name__ == "__main__":
import doctest
doctest.testmod()
| 23 | 0 |
from __future__ import annotations
def SCREAMING_SNAKE_CASE__ ( UpperCamelCase__: list[float] , UpperCamelCase__: list[float] ):
SCREAMING_SNAKE_CASE__ = sorted(numsa + numsa )
SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ = divmod(len(UpperCamelCase__ ) , 2 )
if mod == 1:
return all_numbers[div]
else:
return (all_numbers[div] + all_numbers[div - 1]) / 2
if __name__ == "__main__":
import doctest
doctest.testmod()
_lowerCamelCase = [float(x) for x in input('Enter the elements of first array: ').split()]
_lowerCamelCase = [float(x) for x in input('Enter the elements of second array: ').split()]
print(F'''The median of two arrays is: {median_of_two_arrays(array_a, array_a)}''') | 6 |
import math
from datetime import datetime, timedelta
def _snake_case (__lowercase):
UpperCamelCase_ = year % 19
UpperCamelCase_ = year % 4
UpperCamelCase_ = year % 7
UpperCamelCase_ = math.floor(year / 100)
UpperCamelCase_ = math.floor((13 + 8 * leap_day_inhibits) / 25)
UpperCamelCase_ = leap_day_inhibits / 4
UpperCamelCase_ = (
15 - lunar_orbit_correction + leap_day_inhibits - leap_day_reinstall_number
) % 30
UpperCamelCase_ = (4 + leap_day_inhibits - leap_day_reinstall_number) % 7
# days to be added to March 21
UpperCamelCase_ = (19 * metonic_cycle + secular_moon_shift) % 30
# PHM -> Paschal Full Moon
UpperCamelCase_ = (
2 * julian_leap_year
+ 4 * non_leap_year
+ 6 * days_to_add
+ century_starting_point
) % 7
if days_to_add == 29 and days_from_phm_to_sunday == 6:
return datetime(__lowercase , 4 , 19)
elif days_to_add == 28 and days_from_phm_to_sunday == 6:
return datetime(__lowercase , 4 , 18)
else:
return datetime(__lowercase , 3 , 22) + timedelta(
days=int(days_to_add + days_from_phm_to_sunday))
if __name__ == "__main__":
for year in (1_9_9_4, 2_0_0_0, 2_0_1_0, 2_0_2_1, 2_0_2_3):
snake_case__ : Dict = """will be""" if year > datetime.now().year else """was"""
print(f'Easter in {year} {tense} {gauss_easter(year)}')
| 23 | 0 |
"""simple docstring"""
import argparse
import re
import requests
import torch
# git clone https://github.com/salesforce/BLIP.git
from models.blip import blip_decoder
from models.blip_itm import blip_itm
from models.blip_vqa import blip_vqa
from PIL import Image
from torchvision import transforms
from torchvision.transforms.functional import InterpolationMode
from transformers import (
BertTokenizer,
BlipConfig,
BlipForConditionalGeneration,
BlipForImageTextRetrieval,
BlipForQuestionAnswering,
)
def _snake_case ( _snake_case : Union[str, Any] , _snake_case : List[str] ) -> str:
'''simple docstring'''
_A = 'https://storage.googleapis.com/sfr-vision-language-research/BLIP/demo.jpg'
_A = Image.open(requests.get(_snake_case , stream=_snake_case ).raw ).convert('RGB' )
_A = transforms.Compose(
[
transforms.Resize((image_size, image_size) , interpolation=InterpolationMode.BICUBIC ),
transforms.ToTensor(),
transforms.Normalize((0.48145466, 0.4578275, 0.40821073) , (0.26862954, 0.26130258, 0.27577711) ),
] )
_A = transform(_snake_case ).unsqueeze(0 ).to(_snake_case )
return image
def _snake_case ( _snake_case : List[str] ) -> Dict:
'''simple docstring'''
if "visual_encoder" in key:
_A = re.sub('visual_encoder*' , 'vision_model.encoder' , _snake_case )
if "blocks" in key:
_A = re.sub(R'blocks' , 'layers' , _snake_case )
if "attn" in key:
_A = re.sub(R'attn' , 'self_attn' , _snake_case )
if "norm1" in key:
_A = re.sub(R'norm1' , 'layer_norm1' , _snake_case )
if "norm2" in key:
_A = re.sub(R'norm2' , 'layer_norm2' , _snake_case )
if "encoder.norm" in key:
_A = re.sub(R'encoder.norm' , 'post_layernorm' , _snake_case )
if "encoder.patch_embed.proj" in key:
_A = re.sub(R'encoder.patch_embed.proj' , 'embeddings.patch_embedding' , _snake_case )
if "encoder.pos_embed" in key:
_A = re.sub(R'encoder.pos_embed' , 'embeddings.position_embedding' , _snake_case )
if "encoder.cls_token" in key:
_A = re.sub(R'encoder.cls_token' , 'embeddings.class_embedding' , _snake_case )
if "self_attn" in key:
_A = re.sub(R'self_attn.proj' , 'self_attn.projection' , _snake_case )
return key
@torch.no_grad()
def _snake_case ( _snake_case : Optional[int] , _snake_case : List[str]=None ) -> Any:
'''simple docstring'''
if config_path is not None:
_A = BlipConfig.from_pretrained(_snake_case )
else:
_A = BlipConfig(projection_dim=5_12 , text_config={} , vision_config={} )
_A = BlipForConditionalGeneration(_snake_case ).eval()
_A = 'https://storage.googleapis.com/sfr-vision-language-research/BLIP/models/model_base_capfilt_large.pth'
_A = blip_decoder(pretrained=_snake_case , image_size=3_84 , vit='base' )
_A = pt_model.eval()
_A = pt_model.state_dict()
for key in modified_state_dict.copy():
_A = modified_state_dict.pop(_snake_case )
_A = rename_key(_snake_case )
_A = value
hf_model.load_state_dict(_snake_case )
_A = 3_84
_A = load_demo_image(image_size=_snake_case , device='cpu' )
_A = BertTokenizer.from_pretrained('bert-base-uncased' )
_A = tokenizer(['a picture of'] ).input_ids
_A = hf_model.generate(_snake_case , _snake_case )
assert out[0].tolist() == [3_05_22, 10_37, 38_61, 19_97, 10_37, 24_50, 35_64, 20_06, 19_96, 35_09, 20_07, 20_14, 38_99, 1_02]
_A = hf_model.generate(_snake_case )
assert out[0].tolist() == [3_05_22, 10_37, 24_50, 35_64, 20_06, 19_96, 35_09, 20_07, 20_14, 38_99, 1_02]
if pytorch_dump_folder_path is not None:
hf_model.save_pretrained(_snake_case )
# model_url = 'https://storage.googleapis.com/sfr-vision-language-research/BLIP/models/model_vqa.pth'
_A = (
'https://storage.googleapis.com/sfr-vision-language-research/BLIP/models/model_base_vqa_capfilt_large.pth'
)
_A = blip_vqa(pretrained=_snake_case , image_size=_snake_case , vit='base' )
vqa_model.eval()
_A = vqa_model.state_dict()
for key in modified_state_dict.copy():
_A = modified_state_dict.pop(_snake_case )
_A = rename_key(_snake_case )
_A = value
_A = BlipForQuestionAnswering(_snake_case )
hf_vqa_model.load_state_dict(_snake_case )
_A = ['How many dogs are in this image?']
_A = tokenizer(_snake_case , return_tensors='pt' ).input_ids
_A = hf_vqa_model.generate(_snake_case , _snake_case )
print(tokenizer.decode(answer[0] ) )
assert tokenizer.decode(answer[0] ) == "[UNK] 1 [SEP]"
if pytorch_dump_folder_path is not None:
hf_vqa_model.save_pretrained(pytorch_dump_folder_path + '_vqa' )
_A = 'https://storage.googleapis.com/sfr-vision-language-research/BLIP/models/model_base_retrieval_coco.pth'
_A = blip_itm(pretrained=_snake_case , image_size=_snake_case , vit='base' )
itm_model.eval()
_A = itm_model.state_dict()
for key in modified_state_dict.copy():
_A = modified_state_dict.pop(_snake_case )
_A = rename_key(_snake_case )
_A = value
_A = BlipForImageTextRetrieval(_snake_case )
_A = ['A picture of a woman with a dog sitting in a beach']
_A = tokenizer(
_snake_case , return_tensors='pt' , padding='max_length' , truncation=_snake_case , max_length=35 , ).input_ids
hf_itm_model.load_state_dict(_snake_case )
hf_itm_model.eval()
_A = hf_itm_model(_snake_case , _snake_case , use_itm_head=_snake_case )
_A = hf_itm_model(_snake_case , _snake_case , use_itm_head=_snake_case )
assert out[0].item() == 0.2110687494277954
assert torch.nn.functional.softmax(out_itm[0] , dim=1 )[:, 1].item() == 0.45698845386505127
if pytorch_dump_folder_path is not None:
hf_itm_model.save_pretrained(pytorch_dump_folder_path + '_itm' )
if __name__ == "__main__":
a = argparse.ArgumentParser()
parser.add_argument('''--pytorch_dump_folder_path''', default=None, type=str, help='''Path to the output PyTorch model.''')
parser.add_argument('''--config_path''', default=None, type=str, help='''Path to hf config.json of model to convert''')
a = parser.parse_args()
convert_blip_checkpoint(args.checkpoint_path, args.pytorch_dump_folder_path, args.config_path)
| 7 |
import requests
def _snake_case (__lowercase , __lowercase):
UpperCamelCase_ = {'Content-Type': 'application/json'}
UpperCamelCase_ = requests.post(__lowercase , json={'text': message_body} , headers=__lowercase)
if response.status_code != 200:
UpperCamelCase_ = (
'Request to slack returned an error '
f"""{response.status_code}, the response is:\n{response.text}"""
)
raise ValueError(__lowercase)
if __name__ == "__main__":
# Set the slack url to the one provided by Slack when you create the webhook at
# https://my.slack.com/services/new/incoming-webhook/
send_slack_message("""<YOUR MESSAGE BODY>""", """<SLACK CHANNEL URL>""")
| 23 | 0 |
'''simple docstring'''
from typing import TYPE_CHECKING
from ...utils import (
OptionalDependencyNotAvailable,
_LazyModule,
is_tf_available,
is_tokenizers_available,
is_torch_available,
)
lowercase__ : str = {
'''configuration_convbert''': ['''CONVBERT_PRETRAINED_CONFIG_ARCHIVE_MAP''', '''ConvBertConfig''', '''ConvBertOnnxConfig'''],
'''tokenization_convbert''': ['''ConvBertTokenizer'''],
}
try:
if not is_tokenizers_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
lowercase__ : Tuple = ['''ConvBertTokenizerFast''']
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
lowercase__ : int = [
'''CONVBERT_PRETRAINED_MODEL_ARCHIVE_LIST''',
'''ConvBertForMaskedLM''',
'''ConvBertForMultipleChoice''',
'''ConvBertForQuestionAnswering''',
'''ConvBertForSequenceClassification''',
'''ConvBertForTokenClassification''',
'''ConvBertLayer''',
'''ConvBertModel''',
'''ConvBertPreTrainedModel''',
'''load_tf_weights_in_convbert''',
]
try:
if not is_tf_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
lowercase__ : List[Any] = [
'''TF_CONVBERT_PRETRAINED_MODEL_ARCHIVE_LIST''',
'''TFConvBertForMaskedLM''',
'''TFConvBertForMultipleChoice''',
'''TFConvBertForQuestionAnswering''',
'''TFConvBertForSequenceClassification''',
'''TFConvBertForTokenClassification''',
'''TFConvBertLayer''',
'''TFConvBertModel''',
'''TFConvBertPreTrainedModel''',
]
if TYPE_CHECKING:
from .configuration_convbert import CONVBERT_PRETRAINED_CONFIG_ARCHIVE_MAP, ConvBertConfig, ConvBertOnnxConfig
from .tokenization_convbert import ConvBertTokenizer
try:
if not is_tokenizers_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .tokenization_convbert_fast import ConvBertTokenizerFast
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_convbert import (
CONVBERT_PRETRAINED_MODEL_ARCHIVE_LIST,
ConvBertForMaskedLM,
ConvBertForMultipleChoice,
ConvBertForQuestionAnswering,
ConvBertForSequenceClassification,
ConvBertForTokenClassification,
ConvBertLayer,
ConvBertModel,
ConvBertPreTrainedModel,
load_tf_weights_in_convbert,
)
try:
if not is_tf_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_tf_convbert import (
TF_CONVBERT_PRETRAINED_MODEL_ARCHIVE_LIST,
TFConvBertForMaskedLM,
TFConvBertForMultipleChoice,
TFConvBertForQuestionAnswering,
TFConvBertForSequenceClassification,
TFConvBertForTokenClassification,
TFConvBertLayer,
TFConvBertModel,
TFConvBertPreTrainedModel,
)
else:
import sys
lowercase__ : Union[str, Any] = _LazyModule(__name__, globals()['''__file__'''], _import_structure, module_spec=__spec__) | 8 |
import re
from pathlib import Path
from unittest import TestCase
import pytest
@pytest.mark.integration
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
def _UpperCAmelCase ( self , _UpperCAmelCase ) -> Dict:
with open(_UpperCAmelCase , encoding='utf-8' ) as input_file:
UpperCamelCase_ = re.compile(R'(?!.*\b(?:encoding|rb|w|wb|w+|wb+|ab|ab+)\b)(?<=\s)(open)\((.*)\)' )
UpperCamelCase_ = input_file.read()
UpperCamelCase_ = regexp.search(_UpperCAmelCase )
return match
def _UpperCAmelCase ( self , _UpperCAmelCase ) -> Dict:
with open(_UpperCAmelCase , encoding='utf-8' ) as input_file:
UpperCamelCase_ = re.compile(R'#[^\r\n]*print\(|\"[^\r\n]*print\(|\"\"\".*?print\(.*?\"\"\"|(print\()' , re.DOTALL )
UpperCamelCase_ = input_file.read()
# use `re.finditer` to handle the case where the ignored groups would be matched first by `re.search`
UpperCamelCase_ = regexp.finditer(_UpperCAmelCase )
UpperCamelCase_ = [match for match in matches if match is not None and match.group(1 ) is not None]
return matches[0] if matches else None
def _UpperCAmelCase ( self ) -> List[str]:
UpperCamelCase_ = Path('./datasets' )
UpperCamelCase_ = list(dataset_paths.absolute().glob('**/*.py' ) )
for dataset in dataset_files:
if self._no_encoding_on_file_open(str(_UpperCAmelCase ) ):
raise AssertionError(f"""open(...) must use utf-8 encoding in {dataset}""" )
def _UpperCAmelCase ( self ) -> Optional[int]:
UpperCamelCase_ = Path('./datasets' )
UpperCamelCase_ = list(dataset_paths.absolute().glob('**/*.py' ) )
for dataset in dataset_files:
if self._no_print_statements(str(_UpperCAmelCase ) ):
raise AssertionError(f"""print statement found in {dataset}. Use datasets.logger/logging instead.""" )
| 23 | 0 |
from collections import UserDict
from typing import List, Union
from ..utils import (
add_end_docstrings,
is_tf_available,
is_torch_available,
is_vision_available,
logging,
requires_backends,
)
from .base import PIPELINE_INIT_ARGS, Pipeline
if is_vision_available():
from PIL import Image
from ..image_utils import load_image
if is_torch_available():
from ..models.auto.modeling_auto import MODEL_FOR_ZERO_SHOT_IMAGE_CLASSIFICATION_MAPPING
if is_tf_available():
from ..models.auto.modeling_tf_auto import TF_MODEL_FOR_ZERO_SHOT_IMAGE_CLASSIFICATION_MAPPING
from ..tf_utils import stable_softmax
SCREAMING_SNAKE_CASE__ = logging.get_logger(__name__)
@add_end_docstrings(UpperCAmelCase_ )
class __lowerCAmelCase ( UpperCAmelCase_ ):
"""simple docstring"""
def __init__( self : int , **_snake_case : Dict ):
"""simple docstring"""
super().__init__(**_snake_case )
requires_backends(self , 'vision' )
self.check_model_type(
TF_MODEL_FOR_ZERO_SHOT_IMAGE_CLASSIFICATION_MAPPING
if self.framework == 'tf'
else MODEL_FOR_ZERO_SHOT_IMAGE_CLASSIFICATION_MAPPING )
def __call__( self : Dict , _snake_case : Union[str, List[str], "Image", List["Image"]] , **_snake_case : Optional[int] ):
"""simple docstring"""
return super().__call__(_snake_case , **_snake_case )
def _a ( self : Tuple , **_snake_case : Union[str, Any] ):
"""simple docstring"""
A__ = {}
if "candidate_labels" in kwargs:
A__ = kwargs['candidate_labels']
if "hypothesis_template" in kwargs:
A__ = kwargs['hypothesis_template']
return preprocess_params, {}, {}
def _a ( self : Union[str, Any] , _snake_case : Union[str, Any] , _snake_case : int=None , _snake_case : Optional[Any]="This is a photo of {}." ):
"""simple docstring"""
A__ = load_image(_snake_case )
A__ = self.image_processor(images=[image] , return_tensors=self.framework )
A__ = candidate_labels
A__ = [hypothesis_template.format(_snake_case ) for x in candidate_labels]
A__ = self.tokenizer(_snake_case , return_tensors=self.framework , padding=_snake_case )
A__ = [text_inputs]
return inputs
def _a ( self : Optional[int] , _snake_case : int ):
"""simple docstring"""
A__ = model_inputs.pop('candidate_labels' )
A__ = model_inputs.pop('text_inputs' )
if isinstance(text_inputs[0] , _snake_case ):
A__ = text_inputs[0]
else:
# Batching case.
A__ = text_inputs[0][0]
A__ = self.model(**_snake_case , **_snake_case )
A__ = {
'candidate_labels': candidate_labels,
'logits': outputs.logits_per_image,
}
return model_outputs
def _a ( self : List[Any] , _snake_case : List[Any] ):
"""simple docstring"""
A__ = model_outputs.pop('candidate_labels' )
A__ = model_outputs['logits'][0]
if self.framework == "pt":
A__ = logits.softmax(dim=-1 ).squeeze(-1 )
A__ = probs.tolist()
if not isinstance(_snake_case , _snake_case ):
A__ = [scores]
elif self.framework == "tf":
A__ = stable_softmax(_snake_case , axis=-1 )
A__ = probs.numpy().tolist()
else:
raise ValueError(F'''Unsupported framework: {self.framework}''' )
A__ = [
{'score': score, 'label': candidate_label}
for score, candidate_label in sorted(zip(_snake_case , _snake_case ) , key=lambda _snake_case : -x[0] )
]
return result
| 9 |
import argparse
import random
import joblib
import numpy as np
import torch
from igf.igf import (
SecondaryLearner,
collect_objective_set,
compute_perplexity,
generate_datasets,
load_gpta,
recopy_gpta,
set_seed,
train_secondary_learner,
)
from torch.utils.data import DataLoader, RandomSampler
from transformers import GPTaLMHeadModel
def _snake_case (__lowercase=32 , __lowercase=10 , __lowercase=100 , __lowercase=1026 , __lowercase=True , __lowercase="data/tokenized_stories_train_wikitext103.jbl" , __lowercase="igf_context_pairs.jbl" , ):
set_seed(3)
# generate train_data and objective_set
UpperCamelCase_ , UpperCamelCase_ = generate_datasets(
__lowercase , __lowercase , number=__lowercase , min_len=1026 , trim=__lowercase)
# keeps model same across runs
set_seed(4)
# model, lm_optimizer, lm_scheduler = recopy_gpt2(model, device, max_steps) # store original model weights
# can we train on GPU?
UpperCamelCase_ = torch.device('cuda:0' if torch.cuda.is_available() else 'cpu')
# load pretrained model
UpperCamelCase_ = load_gpta('gpt2').to(__lowercase)
print('computing perplexity on objective set')
UpperCamelCase_ = compute_perplexity(__lowercase , __lowercase , __lowercase).item()
print('perplexity on objective set:' , __lowercase)
# collect igf pairs and save to file demo.jbl
collect_objective_set(__lowercase , __lowercase , __lowercase , __lowercase , __lowercase , __lowercase , __lowercase , __lowercase)
# clean up, delete model and data we don't need anymore
del model, train_data, objective_set
torch.cuda.empty_cache()
def _snake_case (__lowercase , __lowercase=15 , __lowercase=128 , __lowercase=100 , __lowercase="igf_model.pt" , ):
set_seed(42)
# Load pre-trained model
UpperCamelCase_ = GPTaLMHeadModel.from_pretrained('gpt2')
# Initialize secondary learner to use embedding weights of model
UpperCamelCase_ = SecondaryLearner(__lowercase)
# Train secondary learner
UpperCamelCase_ = train_secondary_learner(
__lowercase , __lowercase , max_epochs=__lowercase , batch_size=__lowercase , eval_freq=100 , igf_model_path=__lowercase , )
del model, secondary_learner_train_data
torch.cuda.empty_cache()
return secondary_learner
def _snake_case (__lowercase , __lowercase , __lowercase , __lowercase=32 , __lowercase=1000 , __lowercase=16 , __lowercase=1.0 , __lowercase=recopy_gpta , __lowercase=None , __lowercase=10 , __lowercase="gpt2_finetuned.pt" , ):
UpperCamelCase_ = torch.device('cuda:0' if torch.cuda.is_available() else 'cpu')
UpperCamelCase_ = RandomSampler(__lowercase)
UpperCamelCase_ = DataLoader(__lowercase , sampler=__lowercase)
UpperCamelCase_ = max_steps // (len(__lowercase)) + 1
UpperCamelCase_ = 0
UpperCamelCase_ = torch.zeros((1, context_len) , dtype=torch.long , device=__lowercase)
UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ = recopy_model(__lowercase , __lowercase , __lowercase)
model.train()
if secondary_learner is not None:
secondary_learner.to(__lowercase)
secondary_learner.eval()
UpperCamelCase_ = []
UpperCamelCase_ = 0
UpperCamelCase_ = []
UpperCamelCase_ = []
# Compute the performance of the transformer model at the beginning
UpperCamelCase_ = compute_perplexity(__lowercase , __lowercase , __lowercase)
test_perps.append(__lowercase)
print('Test perplexity, step' , __lowercase , ':' , __lowercase)
for epoch in range(int(__lowercase)):
for step, example in enumerate(__lowercase):
torch.cuda.empty_cache()
UpperCamelCase_ = random.randint(0 , example.size(2) - context_len - 1)
UpperCamelCase_ = example[0, 0, start : start + context_len]
lm_optimizer.zero_grad()
UpperCamelCase_ = model(__lowercase , labels=__lowercase)
UpperCamelCase_ = True
if secondary_learner is not None:
UpperCamelCase_ = secondary_learner.forward(
torch.tensor(__lowercase , dtype=torch.long , device=__lowercase).unsqueeze(0))[0].item()
observed_qs.append(float(__lowercase))
# Here we implement the simple non-constant threshold for the predicted IG(X) value
# We will decay the selectivity of our secondary learner filter from
# 1 standard deviation above average to 1 below average after 10 batches.
if global_step == 10:
UpperCamelCase_ = -1
if predicted_q < threshold:
UpperCamelCase_ = False
# If we passed the filter, add the context to the batch!
if do_backprop:
contexts.append(np.array(context.cpu()))
UpperCamelCase_ = outputs[0]
lm_loss.backward()
examples += 1
del outputs
# Once the batch is filled with enough contexts, backprop on the batch.
if examples == batch_size:
torch.cuda.empty_cache()
UpperCamelCase_ = 0
# Do LM backprop
torch.nn.utils.clip_grad_norm_(model.parameters() , 3.0)
lm_optimizer.step()
lm_scheduler.step() # Update learning rate schedule
global_step += 1
# Compute the performance of the transformer model at this batch
if global_step % eval_interval == 0:
UpperCamelCase_ = compute_perplexity(__lowercase , __lowercase , __lowercase)
test_perps.append(__lowercase)
print('Test perplexity, step' , __lowercase , ':' , __lowercase)
# Break out of the loop after 60 batches
if max_steps > 0 and global_step > 60:
break
if max_steps > 0 and global_step > 60:
break
# save finetuned transformer model
torch.save(model.state_dict() , __lowercase)
torch.cuda.empty_cache()
# Do some cleaning up so we can reinitialize for the next run of this function
del lm_optimizer
del lm_scheduler
return model
def _snake_case ():
UpperCamelCase_ = argparse.ArgumentParser(description='Fine-tune a transformer model with IGF on a language modeling task')
# Required parameters
parser.add_argument(
'--data_dir' , default=__lowercase , type=__lowercase , required=__lowercase , help='The input data dir. Should contain data files for WikiText.' , )
parser.add_argument(
'--model_name_or_path' , default=__lowercase , type=__lowercase , required=__lowercase , help='Path to pretrained model or model identifier from huggingface.co/models' , )
parser.add_argument(
'--data_file' , type=__lowercase , default=__lowercase , help=(
'A jbl file containing tokenized data which can be split as objective dataset, '
'train_dataset and test_dataset.'
) , )
parser.add_argument(
'--igf_data_file' , type=__lowercase , default=__lowercase , help='A jbl file containing the context and information gain pairs to train secondary learner.' , )
parser.add_argument(
'--output_dir' , default=__lowercase , type=__lowercase , required=__lowercase , help='The output directory where the final fine-tuned model is stored.' , )
parser.add_argument(
'--tokenizer_name' , default=__lowercase , type=__lowercase , help='Pretrained tokenizer name or path if not the same as model_name' , )
parser.add_argument('--seed' , type=__lowercase , default=__lowercase , help='A seed for reproducible training.')
parser.add_argument(
'--context_len' , default=32 , type=__lowercase , help=(
'The maximum total input sequence length after tokenization. Sequences longer '
'than this will be truncated, sequences shorter will be padded.'
) , )
parser.add_argument(
'--size_objective_set' , default=100 , type=__lowercase , help='number of articles that are long enough to be used as our objective set' , )
parser.add_argument(
'--eval_freq' , default=100 , type=__lowercase , help='secondary model evaluation is triggered at eval_freq')
parser.add_argument('--max_steps' , default=1000 , type=__lowercase , help='To calculate training epochs')
parser.add_argument(
'--secondary_learner_batch_size' , default=128 , type=__lowercase , help='batch size of training data for secondary learner' , )
parser.add_argument(
'--batch_size' , default=16 , type=__lowercase , help='batch size of training data of language model(gpt2) ')
parser.add_argument(
'--eval_interval' , default=10 , type=__lowercase , help=(
'decay the selectivity of our secondary learner filter from'
'1 standard deviation above average to 1 below average after 10 batches'
) , )
parser.add_argument(
'--number' , default=100 , type=__lowercase , help='The number of examples split to be used as objective_set/test_data')
parser.add_argument(
'--min_len' , default=1026 , type=__lowercase , help='The minimum length of the article to be used as objective set')
parser.add_argument(
'--secondary_learner_max_epochs' , default=15 , type=__lowercase , help='number of epochs to train secondary learner')
parser.add_argument('--trim' , default=__lowercase , type=__lowercase , help='truncate the example if it exceeds context length')
parser.add_argument(
'--threshold' , default=1.0 , type=__lowercase , help=(
'The threshold value used by secondary learner to filter the train_data and allow only'
' informative data as input to the model'
) , )
parser.add_argument('--finetuned_model_name' , default='gpt2_finetuned.pt' , type=__lowercase , help='finetuned_model_name')
parser.add_argument(
'--recopy_model' , default=__lowercase , type=__lowercase , help='Reset the model to the original pretrained GPT-2 weights after each iteration' , )
# function calls
# Collecting *n* pairs of context and information gain(X, IG(X)) for training the secondary learner
generate_n_pairs(
context_len=32 , max_steps=10 , size_objective_set=100 , min_len=1026 , trim=__lowercase , data_file='data/tokenized_stories_train_wikitext103.jbl' , igf_data_file='igf_context_pairs.jbl' , )
# Load train data for secondary learner
UpperCamelCase_ = joblib.load('data/IGF_values.jbl')
# Train secondary learner
UpperCamelCase_ = training_secondary_learner(
__lowercase , secondary_learner_max_epochs=15 , secondary_learner_batch_size=128 , eval_freq=100 , igf_model_path='igf_model.pt' , )
# load pretrained gpt2 model
UpperCamelCase_ = GPTaLMHeadModel.from_pretrained('gpt2')
set_seed(42)
# Generate train and test data to train and evaluate gpt2 model
UpperCamelCase_ , UpperCamelCase_ = generate_datasets(
context_len=32 , file='data/tokenized_stories_train_wikitext103.jbl' , number=100 , min_len=1026 , trim=__lowercase)
# fine-tuning of the gpt2 model using igf (Information Gain Filtration)
finetune(
__lowercase , __lowercase , __lowercase , context_len=32 , max_steps=1000 , batch_size=16 , threshold=1.0 , recopy_model=__lowercase , secondary_learner=__lowercase , eval_interval=10 , finetuned_model_name='gpt2_finetuned.pt' , )
if __name__ == "__main__":
main()
| 23 | 0 |
import os
def _snake_case ( __snake_case = "matrix.txt" ):
with open(os.path.join(os.path.dirname(__snake_case ) , __snake_case ) ) as in_file:
_UpperCamelCase = in_file.read()
_UpperCamelCase = [[int(__snake_case ) for cell in row.split(''',''' )] for row in data.strip().splitlines()]
_UpperCamelCase = [[0 for cell in row] for row in grid]
_UpperCamelCase = len(grid[0] )
_UpperCamelCase = [[0 for i in range(__snake_case )] for j in range(__snake_case )]
_UpperCamelCase = grid[0][0]
for i in range(1 , __snake_case ):
_UpperCamelCase = grid[0][i] + dp[0][i - 1]
for i in range(1 , __snake_case ):
_UpperCamelCase = grid[i][0] + dp[i - 1][0]
for i in range(1 , __snake_case ):
for j in range(1 , __snake_case ):
_UpperCamelCase = grid[i][j] + min(dp[i - 1][j] , dp[i][j - 1] )
return dp[-1][-1]
if __name__ == "__main__":
print(f'{solution() = }')
| 10 |
from __future__ import annotations
import unittest
from transformers import AutoTokenizer, MBartConfig, is_tf_available
from transformers.testing_utils import require_sentencepiece, require_tf, require_tokenizers, slow
from transformers.utils import cached_property
from ...test_configuration_common import ConfigTester
from ...test_modeling_tf_common import TFModelTesterMixin, ids_tensor
from ...test_pipeline_mixin import PipelineTesterMixin
if is_tf_available():
import tensorflow as tf
from transformers import TFAutoModelForSeqaSeqLM, TFMBartForConditionalGeneration, TFMBartModel
@require_tf
class _a :
"""simple docstring"""
A_ = MBartConfig
A_ = {}
A_ = """gelu"""
def __init__( self , _UpperCAmelCase , _UpperCAmelCase=13 , _UpperCAmelCase=7 , _UpperCAmelCase=True , _UpperCAmelCase=False , _UpperCAmelCase=99 , _UpperCAmelCase=32 , _UpperCAmelCase=2 , _UpperCAmelCase=4 , _UpperCAmelCase=37 , _UpperCAmelCase=0.1 , _UpperCAmelCase=0.1 , _UpperCAmelCase=20 , _UpperCAmelCase=2 , _UpperCAmelCase=1 , _UpperCAmelCase=0 , ) -> Union[str, Any]:
UpperCamelCase_ = parent
UpperCamelCase_ = batch_size
UpperCamelCase_ = seq_length
UpperCamelCase_ = is_training
UpperCamelCase_ = use_labels
UpperCamelCase_ = vocab_size
UpperCamelCase_ = hidden_size
UpperCamelCase_ = num_hidden_layers
UpperCamelCase_ = num_attention_heads
UpperCamelCase_ = intermediate_size
UpperCamelCase_ = hidden_dropout_prob
UpperCamelCase_ = attention_probs_dropout_prob
UpperCamelCase_ = max_position_embeddings
UpperCamelCase_ = eos_token_id
UpperCamelCase_ = pad_token_id
UpperCamelCase_ = bos_token_id
def _UpperCAmelCase ( self ) -> List[Any]:
UpperCamelCase_ = ids_tensor([self.batch_size, self.seq_length - 1] , self.vocab_size )
UpperCamelCase_ = tf.expand_dims(tf.constant([self.eos_token_id] * self.batch_size ) , 1 )
UpperCamelCase_ = tf.concat([input_ids, eos_tensor] , axis=1 )
UpperCamelCase_ = ids_tensor([self.batch_size, self.seq_length] , self.vocab_size )
UpperCamelCase_ = self.config_cls(
vocab_size=self.vocab_size , d_model=self.hidden_size , encoder_layers=self.num_hidden_layers , decoder_layers=self.num_hidden_layers , encoder_attention_heads=self.num_attention_heads , decoder_attention_heads=self.num_attention_heads , encoder_ffn_dim=self.intermediate_size , decoder_ffn_dim=self.intermediate_size , dropout=self.hidden_dropout_prob , attention_dropout=self.attention_probs_dropout_prob , max_position_embeddings=self.max_position_embeddings , eos_token_ids=[2] , bos_token_id=self.bos_token_id , pad_token_id=self.pad_token_id , decoder_start_token_id=self.pad_token_id , **self.config_updates , )
UpperCamelCase_ = prepare_mbart_inputs_dict(_UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase )
return config, inputs_dict
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> List[str]:
UpperCamelCase_ = TFMBartModel(config=_UpperCAmelCase ).get_decoder()
UpperCamelCase_ = inputs_dict['input_ids']
UpperCamelCase_ = input_ids[:1, :]
UpperCamelCase_ = inputs_dict['attention_mask'][:1, :]
UpperCamelCase_ = inputs_dict['head_mask']
UpperCamelCase_ = 1
# first forward pass
UpperCamelCase_ = model(_UpperCAmelCase , attention_mask=_UpperCAmelCase , head_mask=_UpperCAmelCase , use_cache=_UpperCAmelCase )
UpperCamelCase_ , UpperCamelCase_ = outputs.to_tuple()
UpperCamelCase_ = past_key_values[1]
def _snake_case (__lowercase , __lowercase , __lowercase , __lowercase=None , __lowercase=None , __lowercase=None , __lowercase=None , __lowercase=None , ):
if attention_mask is None:
UpperCamelCase_ = tf.cast(tf.math.not_equal(__lowercase , config.pad_token_id) , tf.inta)
if decoder_attention_mask is None:
UpperCamelCase_ = tf.concat(
[
tf.ones(decoder_input_ids[:, :1].shape , dtype=tf.inta),
tf.cast(tf.math.not_equal(decoder_input_ids[:, 1:] , config.pad_token_id) , tf.inta),
] , axis=-1 , )
if head_mask is None:
UpperCamelCase_ = tf.ones((config.encoder_layers, config.encoder_attention_heads))
if decoder_head_mask is None:
UpperCamelCase_ = tf.ones((config.decoder_layers, config.decoder_attention_heads))
if cross_attn_head_mask is None:
UpperCamelCase_ = tf.ones((config.decoder_layers, config.decoder_attention_heads))
return {
"input_ids": input_ids,
"decoder_input_ids": decoder_input_ids,
"attention_mask": attention_mask,
"decoder_attention_mask": decoder_attention_mask,
"head_mask": head_mask,
"decoder_head_mask": decoder_head_mask,
"cross_attn_head_mask": cross_attn_head_mask,
}
@require_tf
class _a ( UpperCAmelCase__ , UpperCAmelCase__ , unittest.TestCase ):
"""simple docstring"""
A_ = (TFMBartForConditionalGeneration, TFMBartModel) if is_tf_available() else ()
A_ = (TFMBartForConditionalGeneration,) if is_tf_available() else ()
A_ = (
{
"""conversational""": TFMBartForConditionalGeneration,
"""feature-extraction""": TFMBartModel,
"""summarization""": TFMBartForConditionalGeneration,
"""text2text-generation""": TFMBartForConditionalGeneration,
"""translation""": TFMBartForConditionalGeneration,
}
if is_tf_available()
else {}
)
A_ = True
A_ = False
A_ = False
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase ) -> Tuple:
if pipeline_test_casse_name != "FeatureExtractionPipelineTests":
# Exception encountered when calling layer '...'
return True
return False
def _UpperCAmelCase ( self ) -> Optional[int]:
UpperCamelCase_ = TFMBartModelTester(self )
UpperCamelCase_ = ConfigTester(self , config_class=_UpperCAmelCase )
def _UpperCAmelCase ( self ) -> Optional[int]:
self.config_tester.run_common_tests()
def _UpperCAmelCase ( self ) -> List[Any]:
UpperCamelCase_ = self.model_tester.prepare_config_and_inputs_for_common()
self.model_tester.check_decoder_model_past_large_inputs(*_UpperCAmelCase )
@require_sentencepiece
@require_tokenizers
@require_tf
class _a ( unittest.TestCase ):
"""simple docstring"""
A_ = [
""" UN Chief Says There Is No Military Solution in Syria""",
]
A_ = [
"""Şeful ONU declară că nu există o soluţie militară în Siria""",
]
A_ = """facebook/mbart-large-en-ro"""
@cached_property
def _UpperCAmelCase ( self ) -> Any:
return AutoTokenizer.from_pretrained(self.model_name )
@cached_property
def _UpperCAmelCase ( self ) -> List[str]:
UpperCamelCase_ = TFAutoModelForSeqaSeqLM.from_pretrained(self.model_name )
return model
def _UpperCAmelCase ( self , **_UpperCAmelCase ) -> int:
UpperCamelCase_ = self.translate_src_text(**_UpperCAmelCase )
self.assertListEqual(self.expected_text , _UpperCAmelCase )
def _UpperCAmelCase ( self , **_UpperCAmelCase ) -> List[str]:
UpperCamelCase_ = self.tokenizer(self.src_text , **_UpperCAmelCase , return_tensors='tf' )
UpperCamelCase_ = self.model.generate(
model_inputs.input_ids , attention_mask=model_inputs.attention_mask , num_beams=2 )
UpperCamelCase_ = self.tokenizer.batch_decode(_UpperCAmelCase , skip_special_tokens=_UpperCAmelCase )
return generated_words
@slow
def _UpperCAmelCase ( self ) -> List[Any]:
self._assert_generated_batch_equal_expected()
| 23 | 0 |
'''simple docstring'''
def lowerCAmelCase (__A , __A):
"""simple docstring"""
_a = ''''''
for i in table:
res += inp[i - 1]
return res
def lowerCAmelCase (__A):
"""simple docstring"""
return data[1:] + data[0]
def lowerCAmelCase (__A , __A):
"""simple docstring"""
_a = ''''''
for i in range(len(__A)):
if a[i] == b[i]:
res += "0"
else:
res += "1"
return res
def lowerCAmelCase (__A , __A):
"""simple docstring"""
_a = int('''0b''' + data[0] + data[-1] , 2)
_a = int('''0b''' + data[1:3] , 2)
return bin(s[row][col])[2:]
def lowerCAmelCase (__A , __A , __A , __A , __A):
"""simple docstring"""
_a = message[:4]
_a = message[4:]
_a = apply_table(__A , __A)
_a = xor(__A , __A)
_a = apply_sbox(__A , temp[:4]) # noqa: E741
_a = apply_sbox(__A , temp[4:])
_a = '''0''' * (2 - len(__A)) + l # noqa: E741
_a = '''0''' * (2 - len(__A)) + r
_a = apply_table(l + r , __A)
_a = xor(__A , __A)
return temp + right
if __name__ == "__main__":
lowercase_ = input("Enter 10 bit key: ")
lowercase_ = input("Enter 8 bit message: ")
lowercase_ = [6, 3, 7, 4, 8, 5, 10, 9]
lowercase_ = [3, 5, 2, 7, 4, 10, 1, 9, 8, 6]
lowercase_ = [2, 4, 3, 1]
lowercase_ = [2, 6, 3, 1, 4, 8, 5, 7]
lowercase_ = [4, 1, 3, 5, 7, 2, 8, 6]
lowercase_ = [4, 1, 2, 3, 2, 3, 4, 1]
lowercase_ = [[1, 0, 3, 2], [3, 2, 1, 0], [0, 2, 1, 3], [3, 1, 3, 2]]
lowercase_ = [[0, 1, 2, 3], [2, 0, 1, 3], [3, 0, 1, 0], [2, 1, 0, 3]]
# key generation
lowercase_ = apply_table(key, paa_table)
lowercase_ = temp[:5]
lowercase_ = temp[5:]
lowercase_ = left_shift(left)
lowercase_ = left_shift(right)
lowercase_ = apply_table(left + right, pa_table)
lowercase_ = left_shift(left)
lowercase_ = left_shift(right)
lowercase_ = left_shift(left)
lowercase_ = left_shift(right)
lowercase_ = apply_table(left + right, pa_table)
# encryption
lowercase_ = apply_table(message, IP)
lowercase_ = function(expansion, sa, sa, keya, temp)
lowercase_ = temp[4:] + temp[:4]
lowercase_ = function(expansion, sa, sa, keya, temp)
lowercase_ = apply_table(temp, IP_inv)
print("Cipher text is:", CT)
# decryption
lowercase_ = apply_table(CT, IP)
lowercase_ = function(expansion, sa, sa, keya, temp)
lowercase_ = temp[4:] + temp[:4]
lowercase_ = function(expansion, sa, sa, keya, temp)
lowercase_ = apply_table(temp, IP_inv)
print("Plain text after decypting is:", PT)
| 11 |
def _snake_case (__lowercase):
UpperCamelCase_ = 1
for i in range(1 , num + 1):
fact *= i
return fact
def _snake_case (__lowercase):
UpperCamelCase_ = 0
while number > 0:
UpperCamelCase_ = number % 10
sum_of_digits += last_digit
UpperCamelCase_ = number // 10 # Removing the last_digit from the given number
return sum_of_digits
def _snake_case (__lowercase = 100):
UpperCamelCase_ = factorial(__lowercase)
UpperCamelCase_ = split_and_add(__lowercase)
return result
if __name__ == "__main__":
print(solution(int(input("""Enter the Number: """).strip())))
| 23 | 0 |
def UpperCamelCase ( lowercase_ ) -> bool:
'''simple docstring'''
return credit_card_number.startswith(("""34""", """35""", """37""", """4""", """5""", """6""") )
def UpperCamelCase ( lowercase_ ) -> bool:
'''simple docstring'''
lowercase__ : int = credit_card_number
lowercase__ : Dict = 0
lowercase__ : Dict = len(lowercase_ ) - 2
for i in range(lowercase_ , -1 , -2 ):
# double the value of every second digit
lowercase__ : str = int(cc_number[i] )
digit *= 2
# If doubling of a number results in a two digit number
# i.e greater than 9(e.g., 6 × 2 = 12),
# then add the digits of the product (e.g., 12: 1 + 2 = 3, 15: 1 + 5 = 6),
# to get a single digit number.
if digit > 9:
digit %= 10
digit += 1
lowercase__ : Optional[int] = cc_number[:i] + str(lowercase_ ) + cc_number[i + 1 :]
total += digit
# Sum up the remaining digits
for i in range(len(lowercase_ ) - 1 , -1 , -2 ):
total += int(cc_number[i] )
return total % 10 == 0
def UpperCamelCase ( lowercase_ ) -> bool:
'''simple docstring'''
lowercase__ : str = F'{credit_card_number} is an invalid credit card number because'
if not credit_card_number.isdigit():
print(F'{error_message} it has nonnumerical characters.' )
return False
if not 13 <= len(lowercase_ ) <= 16:
print(F'{error_message} of its length.' )
return False
if not validate_initial_digits(lowercase_ ):
print(F'{error_message} of its first two digits.' )
return False
if not luhn_validation(lowercase_ ):
print(F'{error_message} it fails the Luhn check.' )
return False
print(F'{credit_card_number} is a valid credit card number.' )
return True
if __name__ == "__main__":
import doctest
doctest.testmod()
validate_credit_card_number("""4111111111111111""")
validate_credit_card_number("""32323""")
| 12 |
from typing import Dict, List, Optional, Union
import numpy as np
from ...image_processing_utils import BaseImageProcessor, BatchFeature, get_size_dict
from ...image_transforms import (
center_crop,
get_resize_output_image_size,
normalize,
rescale,
resize,
to_channel_dimension_format,
)
from ...image_utils import (
IMAGENET_STANDARD_MEAN,
IMAGENET_STANDARD_STD,
ChannelDimension,
ImageInput,
PILImageResampling,
is_valid_image,
to_numpy_array,
valid_images,
)
from ...utils import TensorType, is_vision_available, logging
if is_vision_available():
import PIL
snake_case__ : str = logging.get_logger(__name__)
def _snake_case (__lowercase):
if isinstance(__lowercase , (list, tuple)) and isinstance(videos[0] , (list, tuple)) and is_valid_image(videos[0][0]):
return videos
elif isinstance(__lowercase , (list, tuple)) and is_valid_image(videos[0]):
return [videos]
elif is_valid_image(__lowercase):
return [[videos]]
raise ValueError(f"""Could not make batched video from {videos}""")
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
A_ = ["""pixel_values"""]
def __init__( self , _UpperCAmelCase = True , _UpperCAmelCase = None , _UpperCAmelCase = PILImageResampling.BILINEAR , _UpperCAmelCase = True , _UpperCAmelCase = None , _UpperCAmelCase = True , _UpperCAmelCase = 1 / 255 , _UpperCAmelCase = True , _UpperCAmelCase = None , _UpperCAmelCase = None , **_UpperCAmelCase , ) -> None:
super().__init__(**_UpperCAmelCase )
UpperCamelCase_ = size if size is not None else {'shortest_edge': 224}
UpperCamelCase_ = get_size_dict(_UpperCAmelCase , default_to_square=_UpperCAmelCase )
UpperCamelCase_ = crop_size if crop_size is not None else {'height': 224, 'width': 224}
UpperCamelCase_ = get_size_dict(_UpperCAmelCase , param_name='crop_size' )
UpperCamelCase_ = do_resize
UpperCamelCase_ = size
UpperCamelCase_ = do_center_crop
UpperCamelCase_ = crop_size
UpperCamelCase_ = resample
UpperCamelCase_ = do_rescale
UpperCamelCase_ = rescale_factor
UpperCamelCase_ = do_normalize
UpperCamelCase_ = image_mean if image_mean is not None else IMAGENET_STANDARD_MEAN
UpperCamelCase_ = image_std if image_std is not None else IMAGENET_STANDARD_STD
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase = PILImageResampling.BILINEAR , _UpperCAmelCase = None , **_UpperCAmelCase , ) -> np.ndarray:
UpperCamelCase_ = get_size_dict(_UpperCAmelCase , default_to_square=_UpperCAmelCase )
if "shortest_edge" in size:
UpperCamelCase_ = get_resize_output_image_size(_UpperCAmelCase , size['shortest_edge'] , default_to_square=_UpperCAmelCase )
elif "height" in size and "width" in size:
UpperCamelCase_ = (size['height'], size['width'])
else:
raise ValueError(f"""Size must have 'height' and 'width' or 'shortest_edge' as keys. Got {size.keys()}""" )
return resize(_UpperCAmelCase , size=_UpperCAmelCase , resample=_UpperCAmelCase , data_format=_UpperCAmelCase , **_UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase = None , **_UpperCAmelCase , ) -> np.ndarray:
UpperCamelCase_ = get_size_dict(_UpperCAmelCase )
if "height" not in size or "width" not in size:
raise ValueError(f"""Size must have 'height' and 'width' as keys. Got {size.keys()}""" )
return center_crop(_UpperCAmelCase , size=(size['height'], size['width']) , data_format=_UpperCAmelCase , **_UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase = None , **_UpperCAmelCase , ) -> int:
return rescale(_UpperCAmelCase , scale=_UpperCAmelCase , data_format=_UpperCAmelCase , **_UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase = None , **_UpperCAmelCase , ) -> np.ndarray:
return normalize(_UpperCAmelCase , mean=_UpperCAmelCase , std=_UpperCAmelCase , data_format=_UpperCAmelCase , **_UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = ChannelDimension.FIRST , ) -> np.ndarray:
if do_resize and size is None or resample is None:
raise ValueError('Size and resample must be specified if do_resize is True.' )
if do_center_crop and crop_size is None:
raise ValueError('Crop size must be specified if do_center_crop is True.' )
if do_rescale and rescale_factor is None:
raise ValueError('Rescale factor must be specified if do_rescale is True.' )
if do_normalize and (image_mean is None or image_std is None):
raise ValueError('Image mean and std must be specified if do_normalize is True.' )
# All transformations expect numpy arrays.
UpperCamelCase_ = to_numpy_array(_UpperCAmelCase )
if do_resize:
UpperCamelCase_ = self.resize(image=_UpperCAmelCase , size=_UpperCAmelCase , resample=_UpperCAmelCase )
if do_center_crop:
UpperCamelCase_ = self.center_crop(_UpperCAmelCase , size=_UpperCAmelCase )
if do_rescale:
UpperCamelCase_ = self.rescale(image=_UpperCAmelCase , scale=_UpperCAmelCase )
if do_normalize:
UpperCamelCase_ = self.normalize(image=_UpperCAmelCase , mean=_UpperCAmelCase , std=_UpperCAmelCase )
UpperCamelCase_ = to_channel_dimension_format(_UpperCAmelCase , _UpperCAmelCase )
return image
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = ChannelDimension.FIRST , **_UpperCAmelCase , ) -> PIL.Image.Image:
UpperCamelCase_ = do_resize if do_resize is not None else self.do_resize
UpperCamelCase_ = resample if resample is not None else self.resample
UpperCamelCase_ = do_center_crop if do_center_crop is not None else self.do_center_crop
UpperCamelCase_ = do_rescale if do_rescale is not None else self.do_rescale
UpperCamelCase_ = rescale_factor if rescale_factor is not None else self.rescale_factor
UpperCamelCase_ = do_normalize if do_normalize is not None else self.do_normalize
UpperCamelCase_ = image_mean if image_mean is not None else self.image_mean
UpperCamelCase_ = image_std if image_std is not None else self.image_std
UpperCamelCase_ = size if size is not None else self.size
UpperCamelCase_ = get_size_dict(_UpperCAmelCase , default_to_square=_UpperCAmelCase )
UpperCamelCase_ = crop_size if crop_size is not None else self.crop_size
UpperCamelCase_ = get_size_dict(_UpperCAmelCase , param_name='crop_size' )
if not valid_images(_UpperCAmelCase ):
raise ValueError(
'Invalid image type. Must be of type PIL.Image.Image, numpy.ndarray, '
'torch.Tensor, tf.Tensor or jax.ndarray.' )
UpperCamelCase_ = make_batched(_UpperCAmelCase )
UpperCamelCase_ = [
[
self._preprocess_image(
image=_UpperCAmelCase , do_resize=_UpperCAmelCase , size=_UpperCAmelCase , resample=_UpperCAmelCase , do_center_crop=_UpperCAmelCase , crop_size=_UpperCAmelCase , do_rescale=_UpperCAmelCase , rescale_factor=_UpperCAmelCase , do_normalize=_UpperCAmelCase , image_mean=_UpperCAmelCase , image_std=_UpperCAmelCase , data_format=_UpperCAmelCase , )
for img in video
]
for video in videos
]
UpperCamelCase_ = {'pixel_values': videos}
return BatchFeature(data=_UpperCAmelCase , tensor_type=_UpperCAmelCase )
| 23 | 0 |
'''simple docstring'''
from collections import OrderedDict
from typing import Mapping
from ...configuration_utils import PretrainedConfig
from ...onnx import OnnxConfig
from ...utils import logging
A__ : Optional[int] = logging.get_logger(__name__)
A__ : str = {
"""camembert-base""": """https://huggingface.co/camembert-base/resolve/main/config.json""",
"""umberto-commoncrawl-cased-v1""": (
"""https://huggingface.co/Musixmatch/umberto-commoncrawl-cased-v1/resolve/main/config.json"""
),
"""umberto-wikipedia-uncased-v1""": (
"""https://huggingface.co/Musixmatch/umberto-wikipedia-uncased-v1/resolve/main/config.json"""
),
}
class UpperCAmelCase_ (_UpperCAmelCase ):
"""simple docstring"""
lowerCamelCase : str = 'camembert'
def __init__( self , SCREAMING_SNAKE_CASE_=3_05_22 , SCREAMING_SNAKE_CASE_=7_68 , SCREAMING_SNAKE_CASE_=12 , SCREAMING_SNAKE_CASE_=12 , SCREAMING_SNAKE_CASE_=30_72 , SCREAMING_SNAKE_CASE_="gelu" , SCREAMING_SNAKE_CASE_=0.1 , SCREAMING_SNAKE_CASE_=0.1 , SCREAMING_SNAKE_CASE_=5_12 , SCREAMING_SNAKE_CASE_=2 , SCREAMING_SNAKE_CASE_=0.0_2 , SCREAMING_SNAKE_CASE_=1E-12 , SCREAMING_SNAKE_CASE_=1 , SCREAMING_SNAKE_CASE_=0 , SCREAMING_SNAKE_CASE_=2 , SCREAMING_SNAKE_CASE_="absolute" , SCREAMING_SNAKE_CASE_=True , SCREAMING_SNAKE_CASE_=None , **SCREAMING_SNAKE_CASE_ , ) -> Any:
super().__init__(pad_token_id=SCREAMING_SNAKE_CASE_ , bos_token_id=SCREAMING_SNAKE_CASE_ , eos_token_id=SCREAMING_SNAKE_CASE_ , **SCREAMING_SNAKE_CASE_ )
__lowerCamelCase : int = vocab_size
__lowerCamelCase : Optional[int] = hidden_size
__lowerCamelCase : Tuple = num_hidden_layers
__lowerCamelCase : Any = num_attention_heads
__lowerCamelCase : str = hidden_act
__lowerCamelCase : Dict = intermediate_size
__lowerCamelCase : str = hidden_dropout_prob
__lowerCamelCase : Any = attention_probs_dropout_prob
__lowerCamelCase : Tuple = max_position_embeddings
__lowerCamelCase : Dict = type_vocab_size
__lowerCamelCase : List[str] = initializer_range
__lowerCamelCase : Tuple = layer_norm_eps
__lowerCamelCase : Optional[int] = position_embedding_type
__lowerCamelCase : Optional[Any] = use_cache
__lowerCamelCase : int = classifier_dropout
class UpperCAmelCase_ (_UpperCAmelCase ):
"""simple docstring"""
@property
def lowercase_ ( self ) -> Mapping[str, Mapping[int, str]]:
if self.task == "multiple-choice":
__lowerCamelCase : Dict = {0: 'batch', 1: 'choice', 2: 'sequence'}
else:
__lowerCamelCase : str = {0: 'batch', 1: 'sequence'}
return OrderedDict(
[
('input_ids', dynamic_axis),
('attention_mask', dynamic_axis),
] )
| 13 |
# DISCLAIMER: This file is strongly influenced by https://github.com/yang-song/score_sde_pytorch
import math
from dataclasses import dataclass
from typing import Optional, Tuple, Union
import torch
from ..configuration_utils import ConfigMixin, register_to_config
from ..utils import BaseOutput, randn_tensor
from .scheduling_utils import SchedulerMixin, SchedulerOutput
@dataclass
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
A_ = 42
A_ = 42
class _a ( UpperCAmelCase__ , UpperCAmelCase__ ):
"""simple docstring"""
A_ = 1
@register_to_config
def __init__( self , _UpperCAmelCase = 2000 , _UpperCAmelCase = 0.1_5 , _UpperCAmelCase = 0.0_1 , _UpperCAmelCase = 1_3_4_8.0 , _UpperCAmelCase = 1e-5 , _UpperCAmelCase = 1 , ) -> Tuple:
# standard deviation of the initial noise distribution
UpperCamelCase_ = sigma_max
# setable values
UpperCamelCase_ = None
self.set_sigmas(_UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase = None ) -> torch.FloatTensor:
return sample
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase = None , _UpperCAmelCase = None ) -> str:
UpperCamelCase_ = sampling_eps if sampling_eps is not None else self.config.sampling_eps
UpperCamelCase_ = torch.linspace(1 , _UpperCAmelCase , _UpperCAmelCase , device=_UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None ) -> Any:
UpperCamelCase_ = sigma_min if sigma_min is not None else self.config.sigma_min
UpperCamelCase_ = sigma_max if sigma_max is not None else self.config.sigma_max
UpperCamelCase_ = sampling_eps if sampling_eps is not None else self.config.sampling_eps
if self.timesteps is None:
self.set_timesteps(_UpperCAmelCase , _UpperCAmelCase )
UpperCamelCase_ = sigma_min * (sigma_max / sigma_min) ** (self.timesteps / sampling_eps)
UpperCamelCase_ = torch.exp(torch.linspace(math.log(_UpperCAmelCase ) , math.log(_UpperCAmelCase ) , _UpperCAmelCase ) )
UpperCamelCase_ = torch.tensor([sigma_min * (sigma_max / sigma_min) ** t for t in self.timesteps] )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> List[Any]:
return torch.where(
timesteps == 0 , torch.zeros_like(t.to(timesteps.device ) ) , self.discrete_sigmas[timesteps - 1].to(timesteps.device ) , )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase = None , _UpperCAmelCase = True , ) -> Union[SdeVeOutput, Tuple]:
if self.timesteps is None:
raise ValueError(
'`self.timesteps` is not set, you need to run \'set_timesteps\' after creating the scheduler' )
UpperCamelCase_ = timestep * torch.ones(
sample.shape[0] , device=sample.device ) # torch.repeat_interleave(timestep, sample.shape[0])
UpperCamelCase_ = (timestep * (len(self.timesteps ) - 1)).long()
# mps requires indices to be in the same device, so we use cpu as is the default with cuda
UpperCamelCase_ = timesteps.to(self.discrete_sigmas.device )
UpperCamelCase_ = self.discrete_sigmas[timesteps].to(sample.device )
UpperCamelCase_ = self.get_adjacent_sigma(_UpperCAmelCase , _UpperCAmelCase ).to(sample.device )
UpperCamelCase_ = torch.zeros_like(_UpperCAmelCase )
UpperCamelCase_ = (sigma**2 - adjacent_sigma**2) ** 0.5
# equation 6 in the paper: the model_output modeled by the network is grad_x log pt(x)
# also equation 47 shows the analog from SDE models to ancestral sampling methods
UpperCamelCase_ = diffusion.flatten()
while len(diffusion.shape ) < len(sample.shape ):
UpperCamelCase_ = diffusion.unsqueeze(-1 )
UpperCamelCase_ = drift - diffusion**2 * model_output
# equation 6: sample noise for the diffusion term of
UpperCamelCase_ = randn_tensor(
sample.shape , layout=sample.layout , generator=_UpperCAmelCase , device=sample.device , dtype=sample.dtype )
UpperCamelCase_ = sample - drift # subtract because `dt` is a small negative timestep
# TODO is the variable diffusion the correct scaling term for the noise?
UpperCamelCase_ = prev_sample_mean + diffusion * noise # add impact of diffusion field g
if not return_dict:
return (prev_sample, prev_sample_mean)
return SdeVeOutput(prev_sample=_UpperCAmelCase , prev_sample_mean=_UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase = None , _UpperCAmelCase = True , ) -> Union[SchedulerOutput, Tuple]:
if self.timesteps is None:
raise ValueError(
'`self.timesteps` is not set, you need to run \'set_timesteps\' after creating the scheduler' )
# For small batch sizes, the paper "suggest replacing norm(z) with sqrt(d), where d is the dim. of z"
# sample noise for correction
UpperCamelCase_ = randn_tensor(sample.shape , layout=sample.layout , generator=_UpperCAmelCase ).to(sample.device )
# compute step size from the model_output, the noise, and the snr
UpperCamelCase_ = torch.norm(model_output.reshape(model_output.shape[0] , -1 ) , dim=-1 ).mean()
UpperCamelCase_ = torch.norm(noise.reshape(noise.shape[0] , -1 ) , dim=-1 ).mean()
UpperCamelCase_ = (self.config.snr * noise_norm / grad_norm) ** 2 * 2
UpperCamelCase_ = step_size * torch.ones(sample.shape[0] ).to(sample.device )
# self.repeat_scalar(step_size, sample.shape[0])
# compute corrected sample: model_output term and noise term
UpperCamelCase_ = step_size.flatten()
while len(step_size.shape ) < len(sample.shape ):
UpperCamelCase_ = step_size.unsqueeze(-1 )
UpperCamelCase_ = sample + step_size * model_output
UpperCamelCase_ = prev_sample_mean + ((step_size * 2) ** 0.5) * noise
if not return_dict:
return (prev_sample,)
return SchedulerOutput(prev_sample=_UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , ) -> torch.FloatTensor:
# Make sure sigmas and timesteps have the same device and dtype as original_samples
UpperCamelCase_ = timesteps.to(original_samples.device )
UpperCamelCase_ = self.discrete_sigmas.to(original_samples.device )[timesteps]
UpperCamelCase_ = (
noise * sigmas[:, None, None, None]
if noise is not None
else torch.randn_like(_UpperCAmelCase ) * sigmas[:, None, None, None]
)
UpperCamelCase_ = noise + original_samples
return noisy_samples
def __len__( self ) -> Optional[int]:
return self.config.num_train_timesteps
| 23 | 0 |
import json
import os
from datetime import date
from pathlib import Path
from tabulate import DataRow, TableFormat, tabulate
a__ = TableFormat(
lineabove=None,
linebelowheader=None,
linebetweenrows=None,
linebelow=None,
headerrow=DataRow('''''', '''|''', '''|'''),
datarow=DataRow('''''', '''|''', '''|'''),
padding=1,
with_header_hide=None,
)
a__ = []
a__ = []
a__ = {'''type''': '''section''', '''text''': {'''type''': '''plain_text''', '''text''': '''No failed tests! 🤗''', '''emoji''': True}}
a__ = [
{
'''type''': '''header''',
'''text''': {
'''type''': '''plain_text''',
'''text''': f'''🤗 Accelerate nightly {os.environ.get("TEST_TYPE", "")} test results''',
'''emoji''': True,
},
}
]
a__ = 0
for log in Path().glob('''*.log'''):
a__ = 0
with open(log, '''r''') as f:
for line in f:
a__ = json.loads(line)
if line.get('''nodeid''', '''''') != "":
a__ = line['''nodeid''']
if line.get('''duration''', None) is not None:
a__ = f'''{line["duration"]:.4f}'''
if line.get('''outcome''', '''''') == "failed":
section_num_failed += 1
failed.append([test, duration, log.name.split('''_''')[0]])
total_num_failed += 1
group_info.append([str(log), section_num_failed, failed])
a__ = []
log.unlink()
a__ = ''''''
a__ = []
if total_num_failed > 0:
for name, num_failed, failed_tests in group_info:
if num_failed > 0:
if num_failed == 1:
message += f"*{name[1:]}: {num_failed} failed test*\n"
else:
message += f"*{name[1:]}: {num_failed} failed tests*\n"
a__ = []
a__ = {}
for test in failed_tests:
a__ = test[0].split('''::''')
a__ = data[0].split('''/''')[-1]
if data[0] not in filesafailed:
a__ = [data[1:]]
else:
filesafailed[data[0]] += [data[1:]]
failed_table.append(data)
a__ = [test[0] for test in failed_table]
a__ = list(set(files))
# Count number of instances in failed_tests
a__ = []
for file in individual_files:
table.append([file, len(filesafailed[file])])
a__ = tabulate(
table,
headers=['''Test Location''', '''Num Failed'''],
tablefmt=hf_table_format,
stralign='''right''',
)
message += f"\n```\n{failed_table}\n```"
all_filesafailed.append(filesafailed)
if len(message) > 3000:
a__ = '''Too many failed tests, please see the full report in the Action results.'''
a__ = len(err) + 10
a__ = message[: 3000 - offset] + f'''\n...\n```\n{err}'''
print(f'''### {message}''')
else:
a__ = '''No failed tests! 🤗'''
print(f'''## {message}''')
payload.append(no_error_payload)
if os.environ.get('''TEST_TYPE''', '''''') != "":
from slack_sdk import WebClient
a__ = WebClient(token=os.environ['''SLACK_API_TOKEN'''])
if message != "No failed tests! 🤗":
a__ = {
'''type''': '''section''',
'''text''': {
'''type''': '''mrkdwn''',
'''text''': message,
},
}
payload.append(md_report)
a__ = {
'''type''': '''section''',
'''text''': {
'''type''': '''mrkdwn''',
'''text''': '''*For more details:*''',
},
'''accessory''': {
'''type''': '''button''',
'''text''': {
'''type''': '''plain_text''',
'''text''': '''Check Action results''',
'''emoji''': True,
},
'''url''': f'''https://github.com/{os.environ["GITHUB_REPOSITORY"]}/actions/runs/{os.environ["GITHUB_RUN_ID"]}''',
},
}
payload.append(action_button)
a__ = {
'''type''': '''context''',
'''elements''': [
{
'''type''': '''plain_text''',
'''text''': f'''Nightly {os.environ.get("TEST_TYPE")} test results for {date.today()}''',
}
],
}
payload.append(date_report)
a__ = client.chat_postMessage(channel='''#accelerate-ci-daily''', text=message, blocks=payload)
a__ = response.data['''ts''']
for failed_file in all_filesafailed:
for test_location, test_failures in failed_file.items():
# Keep only the first instance of the test name
a__ = ''''''
for i, row in enumerate(test_failures):
if row[0] != test_class:
a__ = row[0]
else:
a__ = ''''''
a__ = {
'''type''': '''section''',
'''text''': {
'''type''': '''mrkdwn''',
'''text''': f'''Test location: {test_location}\n```\n{tabulate(test_failures, headers=["Class", "Test"], tablefmt=hf_table_format, stralign="right")}\n```''',
},
}
client.chat_postMessage(
channel='''#accelerate-ci-daily''',
thread_ts=ts,
blocks=[payload],
)
| 14 |
from typing import TYPE_CHECKING
from ...utils import OptionalDependencyNotAvailable, _LazyModule, is_torch_available
snake_case__ : Optional[int] = {
"""configuration_pegasus_x""": ["""PEGASUS_X_PRETRAINED_CONFIG_ARCHIVE_MAP""", """PegasusXConfig"""],
}
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
snake_case__ : Dict = [
"""PEGASUS_X_PRETRAINED_MODEL_ARCHIVE_LIST""",
"""PegasusXForConditionalGeneration""",
"""PegasusXModel""",
"""PegasusXPreTrainedModel""",
]
if TYPE_CHECKING:
from .configuration_pegasus_x import PEGASUS_X_PRETRAINED_CONFIG_ARCHIVE_MAP, PegasusXConfig
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_pegasus_x import (
PEGASUS_X_PRETRAINED_MODEL_ARCHIVE_LIST,
PegasusXForConditionalGeneration,
PegasusXModel,
PegasusXPreTrainedModel,
)
else:
import sys
snake_case__ : Optional[Any] = _LazyModule(__name__, globals()["""__file__"""], _import_structure, module_spec=__spec__)
| 23 | 0 |
import json
import os
import unittest
from transformers import MgpstrTokenizer
from transformers.models.mgp_str.tokenization_mgp_str import VOCAB_FILES_NAMES
from transformers.testing_utils import require_tokenizers
from ...test_tokenization_common import TokenizerTesterMixin
@require_tokenizers
class A ( UpperCAmelCase__ , unittest.TestCase ):
'''simple docstring'''
A__ = MgpstrTokenizer
A__ = False
A__ = {}
A__ = False
def lowerCamelCase__ (self : Union[str, Any] ) -> Optional[Any]:
"""simple docstring"""
super().setUp()
# fmt: off
lowercase__ = ["""[GO]""", """[s]""", """0""", """1""", """2""", """3""", """4""", """5""", """6""", """7""", """8""", """9""", """a""", """b""", """c""", """d""", """e""", """f""", """g""", """h""", """i""", """j""", """k""", """l""", """m""", """n""", """o""", """p""", """q""", """r""", """s""", """t""", """u""", """v""", """w""", """x""", """y""", """z"""]
# fmt: on
lowercase__ = dict(zip(_UpperCAmelCase , range(len(_UpperCAmelCase ) ) ) )
lowercase__ = os.path.join(self.tmpdirname , VOCAB_FILES_NAMES["""vocab_file"""] )
with open(self.vocab_file , """w""" , encoding="""utf-8""" ) as fp:
fp.write(json.dumps(_UpperCAmelCase ) + """\n""" )
def lowerCamelCase__ (self : Dict , **_UpperCAmelCase : int ) -> Dict:
"""simple docstring"""
return MgpstrTokenizer.from_pretrained(self.tmpdirname , **_UpperCAmelCase )
def lowerCamelCase__ (self : Optional[int] , _UpperCAmelCase : Any ) -> str:
"""simple docstring"""
lowercase__ = """tester"""
lowercase__ = """tester"""
return input_text, output_text
@unittest.skip("""MGP-STR always lower cases letters.""" )
def lowerCamelCase__ (self : Union[str, Any] ) -> Tuple:
"""simple docstring"""
pass
def lowerCamelCase__ (self : List[str] ) -> Optional[int]:
"""simple docstring"""
lowercase__ = self.get_tokenizers(do_lower_case=_UpperCAmelCase )
for tokenizer in tokenizers:
with self.subTest(f'''{tokenizer.__class__.__name__}''' ):
lowercase__ = """[SPECIAL_TOKEN]"""
tokenizer.add_special_tokens({"""cls_token""": special_token} )
lowercase__ = tokenizer.encode([special_token] , add_special_tokens=_UpperCAmelCase )
self.assertEqual(len(_UpperCAmelCase ) , 1 )
lowercase__ = tokenizer.decode(_UpperCAmelCase , skip_special_tokens=_UpperCAmelCase )
self.assertTrue(special_token not in decoded )
def lowerCamelCase__ (self : Any ) -> List[str]:
"""simple docstring"""
lowercase__ = self.get_tokenizers()
for tokenizer in tokenizers:
with self.subTest(f'''{tokenizer.__class__.__name__}''' ):
lowercase__ , lowercase__ = self.get_input_output_texts(_UpperCAmelCase )
lowercase__ = tokenizer.tokenize(_UpperCAmelCase )
lowercase__ = tokenizer.convert_tokens_to_ids(_UpperCAmelCase )
lowercase__ = tokenizer.encode(_UpperCAmelCase , add_special_tokens=_UpperCAmelCase )
self.assertListEqual(_UpperCAmelCase , _UpperCAmelCase )
lowercase__ = tokenizer.convert_ids_to_tokens(_UpperCAmelCase )
self.assertNotEqual(len(_UpperCAmelCase ) , 0 )
lowercase__ = tokenizer.decode(_UpperCAmelCase )
self.assertIsInstance(_UpperCAmelCase , _UpperCAmelCase )
self.assertEqual(text_a.replace(""" """ , """""" ) , _UpperCAmelCase )
@unittest.skip("""MGP-STR tokenizer only handles one sequence.""" )
def lowerCamelCase__ (self : int ) -> Any:
"""simple docstring"""
pass
@unittest.skip("""inputs cannot be pretokenized in MgpstrTokenizer""" )
def lowerCamelCase__ (self : Optional[Any] ) -> str:
"""simple docstring"""
pass
| 15 |
import datasets
from .evaluate import evaluate
snake_case__ : int = """\
@article{hendrycks2021cuad,
title={CUAD: An Expert-Annotated NLP Dataset for Legal Contract Review},
author={Dan Hendrycks and Collin Burns and Anya Chen and Spencer Ball},
journal={arXiv preprint arXiv:2103.06268},
year={2021}
}
"""
snake_case__ : Union[str, Any] = """
This metric wrap the official scoring script for version 1 of the Contract
Understanding Atticus Dataset (CUAD).
Contract Understanding Atticus Dataset (CUAD) v1 is a corpus of more than 13,000 labels in 510
commercial legal contracts that have been manually labeled to identify 41 categories of important
clauses that lawyers look for when reviewing contracts in connection with corporate transactions.
"""
snake_case__ : Any = """
Computes CUAD scores (EM, F1, AUPR, Precision@80%Recall, and Precision@90%Recall).
Args:
predictions: List of question-answers dictionaries with the following key-values:
- 'id': id of the question-answer pair as given in the references (see below)
- 'prediction_text': list of possible texts for the answer, as a list of strings
depending on a threshold on the confidence probability of each prediction.
references: List of question-answers dictionaries with the following key-values:
- 'id': id of the question-answer pair (see above),
- 'answers': a Dict in the CUAD dataset format
{
'text': list of possible texts for the answer, as a list of strings
'answer_start': list of start positions for the answer, as a list of ints
}
Note that answer_start values are not taken into account to compute the metric.
Returns:
'exact_match': Exact match (the normalized answer exactly match the gold answer)
'f1': The F-score of predicted tokens versus the gold answer
'aupr': Area Under the Precision-Recall curve
'prec_at_80_recall': Precision at 80% recall
'prec_at_90_recall': Precision at 90% recall
Examples:
>>> predictions = [{'prediction_text': ['The seller:', 'The buyer/End-User: Shenzhen LOHAS Supply Chain Management Co., Ltd.'], 'id': 'LohaCompanyltd_20191209_F-1_EX-10.16_11917878_EX-10.16_Supply Agreement__Parties'}]
>>> references = [{'answers': {'answer_start': [143, 49], 'text': ['The seller:', 'The buyer/End-User: Shenzhen LOHAS Supply Chain Management Co., Ltd.']}, 'id': 'LohaCompanyltd_20191209_F-1_EX-10.16_11917878_EX-10.16_Supply Agreement__Parties'}]
>>> cuad_metric = datasets.load_metric(\"cuad\")
>>> results = cuad_metric.compute(predictions=predictions, references=references)
>>> print(results)
{'exact_match': 100.0, 'f1': 100.0, 'aupr': 0.0, 'prec_at_80_recall': 1.0, 'prec_at_90_recall': 1.0}
"""
@datasets.utils.file_utils.add_start_docstrings(_DESCRIPTION , _KWARGS_DESCRIPTION )
class _a ( datasets.Metric ):
"""simple docstring"""
def _UpperCAmelCase ( self ) -> Optional[int]:
return datasets.MetricInfo(
description=_DESCRIPTION , citation=_CITATION , inputs_description=_KWARGS_DESCRIPTION , features=datasets.Features(
{
'predictions': {
'id': datasets.Value('string' ),
'prediction_text': datasets.features.Sequence(datasets.Value('string' ) ),
},
'references': {
'id': datasets.Value('string' ),
'answers': datasets.features.Sequence(
{
'text': datasets.Value('string' ),
'answer_start': datasets.Value('int32' ),
} ),
},
} ) , codebase_urls=['https://www.atticusprojectai.org/cuad'] , reference_urls=['https://www.atticusprojectai.org/cuad'] , )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> Dict:
UpperCamelCase_ = {prediction['id']: prediction['prediction_text'] for prediction in predictions}
UpperCamelCase_ = [
{
'paragraphs': [
{
'qas': [
{
'answers': [{'text': answer_text} for answer_text in ref['answers']['text']],
'id': ref['id'],
}
for ref in references
]
}
]
}
]
UpperCamelCase_ = evaluate(dataset=_UpperCAmelCase , predictions=_UpperCAmelCase )
return score
| 23 | 0 |
import json
import os
from typing import Dict, List, Optional, Tuple
import regex as re
from ...tokenization_utils import PreTrainedTokenizer
from ...utils import logging
__A : List[Any] = logging.get_logger(__name__)
__A : Dict = {
'vocab_file': 'vocab.json',
'merges_file': 'merges.txt',
'tokenizer_config_file': 'tokenizer_config.json',
}
__A : Optional[int] = {
'vocab_file': {
'facebook/blenderbot_small-90M': 'https://huggingface.co/facebook/blenderbot_small-90M/resolve/main/vocab.json'
},
'merges_file': {
'facebook/blenderbot_small-90M': 'https://huggingface.co/facebook/blenderbot_small-90M/resolve/main/merges.txt'
},
'tokenizer_config_file': {
'facebook/blenderbot_small-90M': (
'https://huggingface.co/facebook/blenderbot_small-90M/resolve/main/tokenizer_config.json'
)
},
}
__A : Any = {'facebook/blenderbot_small-90M': 5_1_2}
def __a ( A__ : Dict ):
SCREAMING_SNAKE_CASE = set()
SCREAMING_SNAKE_CASE = word[0]
for char in word[1:]:
pairs.add((prev_char, char) )
SCREAMING_SNAKE_CASE = char
SCREAMING_SNAKE_CASE = set(A__ )
return pairs
class _SCREAMING_SNAKE_CASE ( __snake_case ):
'''simple docstring'''
lowerCamelCase__ = VOCAB_FILES_NAMES
lowerCamelCase__ = PRETRAINED_VOCAB_FILES_MAP
lowerCamelCase__ = PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES
lowerCamelCase__ = ["input_ids", "attention_mask"]
def __init__( self : Dict , __lowerCamelCase : List[Any] , __lowerCamelCase : int , __lowerCamelCase : str="__start__" , __lowerCamelCase : List[str]="__end__" , __lowerCamelCase : Any="__unk__" , __lowerCamelCase : Optional[Any]="__null__" , **__lowerCamelCase : Any , ):
super().__init__(unk_token=__lowerCamelCase , bos_token=__lowerCamelCase , eos_token=__lowerCamelCase , pad_token=__lowerCamelCase , **__lowerCamelCase )
with open(__lowerCamelCase , encoding="utf-8" ) as vocab_handle:
SCREAMING_SNAKE_CASE = json.load(__lowerCamelCase )
SCREAMING_SNAKE_CASE = {v: k for k, v in self.encoder.items()}
with open(__lowerCamelCase , encoding="utf-8" ) as merges_handle:
SCREAMING_SNAKE_CASE = merges_handle.read().split("\n" )[1:-1]
SCREAMING_SNAKE_CASE = [tuple(merge.split() ) for merge in merges]
SCREAMING_SNAKE_CASE = dict(zip(__lowerCamelCase , range(len(__lowerCamelCase ) ) ) )
SCREAMING_SNAKE_CASE = {}
@property
def _snake_case ( self : Optional[int] ):
return len(self.encoder )
def _snake_case ( self : Dict ):
return dict(self.encoder , **self.added_tokens_encoder )
def _snake_case ( self : List[Any] , __lowerCamelCase : str ):
if token in self.cache:
return self.cache[token]
SCREAMING_SNAKE_CASE = re.sub("([.,!?()])" , r" \1" , __lowerCamelCase )
SCREAMING_SNAKE_CASE = re.sub("(')" , r" \1 " , __lowerCamelCase )
SCREAMING_SNAKE_CASE = re.sub(r"\s{2,}" , " " , __lowerCamelCase )
if "\n" in token:
SCREAMING_SNAKE_CASE = token.replace("\n" , " __newln__" )
SCREAMING_SNAKE_CASE = token.split(" " )
SCREAMING_SNAKE_CASE = []
for token in tokens:
if not len(__lowerCamelCase ):
continue
SCREAMING_SNAKE_CASE = token.lower()
SCREAMING_SNAKE_CASE = tuple(__lowerCamelCase )
SCREAMING_SNAKE_CASE = tuple(list(word[:-1] ) + [word[-1] + "</w>"] )
SCREAMING_SNAKE_CASE = get_pairs(__lowerCamelCase )
if not pairs:
words.append(__lowerCamelCase )
continue
while True:
SCREAMING_SNAKE_CASE = min(__lowerCamelCase , key=lambda __lowerCamelCase : self.bpe_ranks.get(__lowerCamelCase , float("inf" ) ) )
if bigram not in self.bpe_ranks:
break
SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE = bigram
SCREAMING_SNAKE_CASE = []
SCREAMING_SNAKE_CASE = 0
while i < len(__lowerCamelCase ):
try:
SCREAMING_SNAKE_CASE = word.index(__lowerCamelCase , __lowerCamelCase )
new_word.extend(word[i:j] )
SCREAMING_SNAKE_CASE = j
except ValueError:
new_word.extend(word[i:] )
break
if word[i] == first and i < len(__lowerCamelCase ) - 1 and word[i + 1] == second:
new_word.append(first + second )
i += 2
else:
new_word.append(word[i] )
i += 1
SCREAMING_SNAKE_CASE = tuple(__lowerCamelCase )
SCREAMING_SNAKE_CASE = new_word
if len(__lowerCamelCase ) == 1:
break
else:
SCREAMING_SNAKE_CASE = get_pairs(__lowerCamelCase )
SCREAMING_SNAKE_CASE = "@@ ".join(__lowerCamelCase )
SCREAMING_SNAKE_CASE = word[:-4]
SCREAMING_SNAKE_CASE = word
words.append(__lowerCamelCase )
return " ".join(__lowerCamelCase )
def _snake_case ( self : str , __lowerCamelCase : str ):
SCREAMING_SNAKE_CASE = []
SCREAMING_SNAKE_CASE = re.findall(r"\S+\n?" , __lowerCamelCase )
for token in words:
split_tokens.extend(list(self.bpe(__lowerCamelCase ).split(" " ) ) )
return split_tokens
def _snake_case ( self : Union[str, Any] , __lowerCamelCase : str ):
SCREAMING_SNAKE_CASE = token.lower()
return self.encoder.get(__lowerCamelCase , self.encoder.get(self.unk_token ) )
def _snake_case ( self : Union[str, Any] , __lowerCamelCase : int ):
return self.decoder.get(__lowerCamelCase , self.unk_token )
def _snake_case ( self : int , __lowerCamelCase : List[str] ):
SCREAMING_SNAKE_CASE = " ".join(__lowerCamelCase ).replace("@@ " , "" ).strip()
return out_string
def _snake_case ( self : Tuple , __lowerCamelCase : str , __lowerCamelCase : Optional[str] = None ):
if not os.path.isdir(__lowerCamelCase ):
logger.error(f"Vocabulary path ({save_directory}) should be a directory" )
return
SCREAMING_SNAKE_CASE = os.path.join(
__lowerCamelCase , (filename_prefix + "-" if filename_prefix else "") + VOCAB_FILES_NAMES["vocab_file"] )
SCREAMING_SNAKE_CASE = os.path.join(
__lowerCamelCase , (filename_prefix + "-" if filename_prefix else "") + VOCAB_FILES_NAMES["merges_file"] )
with open(__lowerCamelCase , "w" , encoding="utf-8" ) as f:
f.write(json.dumps(self.encoder , indent=2 , sort_keys=__lowerCamelCase , ensure_ascii=__lowerCamelCase ) + "\n" )
SCREAMING_SNAKE_CASE = 0
with open(__lowerCamelCase , "w" , encoding="utf-8" ) as writer:
writer.write("#version: 0.2\n" )
for bpe_tokens, token_index in sorted(self.bpe_ranks.items() , key=lambda __lowerCamelCase : kv[1] ):
if index != token_index:
logger.warning(
f"Saving vocabulary to {merge_file}: BPE merge indices are not consecutive."
" Please check that the tokenizer is not corrupted!" )
SCREAMING_SNAKE_CASE = token_index
writer.write(" ".join(__lowerCamelCase ) + "\n" )
index += 1
return vocab_file, merge_file | 16 |
import os
import tempfile
from functools import partial
from unittest import TestCase
from unittest.mock import patch
import datasets
import datasets.config
from .utils import require_beam
class _a ( datasets.BeamBasedBuilder ):
"""simple docstring"""
def _UpperCAmelCase ( self ) -> List[str]:
return datasets.DatasetInfo(
features=datasets.Features({'content': datasets.Value('string' )} ) , supervised_keys=_UpperCAmelCase , )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> Optional[Any]:
return [datasets.SplitGenerator(name=datasets.Split.TRAIN , gen_kwargs={'examples': get_test_dummy_examples()} )]
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> Tuple:
import apache_beam as beam
return pipeline | "Load Examples" >> beam.Create(_UpperCAmelCase )
class _a ( datasets.BeamBasedBuilder ):
"""simple docstring"""
def _UpperCAmelCase ( self ) -> Any:
return datasets.DatasetInfo(
features=datasets.Features({'a': datasets.Sequence({'b': datasets.Value('string' )} )} ) , supervised_keys=_UpperCAmelCase , )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> Tuple:
return [
datasets.SplitGenerator(name=datasets.Split.TRAIN , gen_kwargs={'examples': get_test_nested_examples()} )
]
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> Union[str, Any]:
import apache_beam as beam
return pipeline | "Load Examples" >> beam.Create(_UpperCAmelCase )
def _snake_case ():
return [(i, {"content": content}) for i, content in enumerate(['foo', 'bar', 'foobar'])]
def _snake_case ():
return [(i, {"a": {"b": [content]}}) for i, content in enumerate(['foo', 'bar', 'foobar'])]
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
@require_beam
def _UpperCAmelCase ( self ) -> Dict:
UpperCamelCase_ = len(get_test_dummy_examples() )
with tempfile.TemporaryDirectory() as tmp_cache_dir:
UpperCamelCase_ = DummyBeamDataset(cache_dir=_UpperCAmelCase , beam_runner='DirectRunner' )
builder.download_and_prepare()
self.assertTrue(
os.path.exists(
os.path.join(_UpperCAmelCase , builder.name , 'default' , '0.0.0' , f"""{builder.name}-train.arrow""" ) ) )
self.assertDictEqual(builder.info.features , datasets.Features({'content': datasets.Value('string' )} ) )
UpperCamelCase_ = builder.as_dataset()
self.assertEqual(dset['train'].num_rows , _UpperCAmelCase )
self.assertEqual(dset['train'].info.splits['train'].num_examples , _UpperCAmelCase )
self.assertDictEqual(dset['train'][0] , get_test_dummy_examples()[0][1] )
self.assertDictEqual(
dset['train'][expected_num_examples - 1] , get_test_dummy_examples()[expected_num_examples - 1][1] )
self.assertTrue(
os.path.exists(os.path.join(_UpperCAmelCase , builder.name , 'default' , '0.0.0' , 'dataset_info.json' ) ) )
del dset
@require_beam
def _UpperCAmelCase ( self ) -> List[str]:
import apache_beam as beam
UpperCamelCase_ = beam.io.parquetio.WriteToParquet
UpperCamelCase_ = len(get_test_dummy_examples() )
with tempfile.TemporaryDirectory() as tmp_cache_dir:
UpperCamelCase_ = DummyBeamDataset(cache_dir=_UpperCAmelCase , beam_runner='DirectRunner' )
with patch('apache_beam.io.parquetio.WriteToParquet' ) as write_parquet_mock:
UpperCamelCase_ = partial(_UpperCAmelCase , num_shards=2 )
builder.download_and_prepare()
self.assertTrue(
os.path.exists(
os.path.join(
_UpperCAmelCase , builder.name , 'default' , '0.0.0' , f"""{builder.name}-train-00000-of-00002.arrow""" ) ) )
self.assertTrue(
os.path.exists(
os.path.join(
_UpperCAmelCase , builder.name , 'default' , '0.0.0' , f"""{builder.name}-train-00000-of-00002.arrow""" ) ) )
self.assertDictEqual(builder.info.features , datasets.Features({'content': datasets.Value('string' )} ) )
UpperCamelCase_ = builder.as_dataset()
self.assertEqual(dset['train'].num_rows , _UpperCAmelCase )
self.assertEqual(dset['train'].info.splits['train'].num_examples , _UpperCAmelCase )
# Order is not preserved when sharding, so we just check that all the elements are there
self.assertListEqual(sorted(dset['train']['content'] ) , sorted(['foo', 'bar', 'foobar'] ) )
self.assertTrue(
os.path.exists(os.path.join(_UpperCAmelCase , builder.name , 'default' , '0.0.0' , 'dataset_info.json' ) ) )
del dset
@require_beam
def _UpperCAmelCase ( self ) -> Any:
with tempfile.TemporaryDirectory() as tmp_cache_dir:
UpperCamelCase_ = DummyBeamDataset(cache_dir=_UpperCAmelCase )
self.assertRaises(datasets.builder.MissingBeamOptions , builder.download_and_prepare )
@require_beam
def _UpperCAmelCase ( self ) -> Union[str, Any]:
UpperCamelCase_ = len(get_test_nested_examples() )
with tempfile.TemporaryDirectory() as tmp_cache_dir:
UpperCamelCase_ = NestedBeamDataset(cache_dir=_UpperCAmelCase , beam_runner='DirectRunner' )
builder.download_and_prepare()
self.assertTrue(
os.path.exists(
os.path.join(_UpperCAmelCase , builder.name , 'default' , '0.0.0' , f"""{builder.name}-train.arrow""" ) ) )
self.assertDictEqual(
builder.info.features , datasets.Features({'a': datasets.Sequence({'b': datasets.Value('string' )} )} ) )
UpperCamelCase_ = builder.as_dataset()
self.assertEqual(dset['train'].num_rows , _UpperCAmelCase )
self.assertEqual(dset['train'].info.splits['train'].num_examples , _UpperCAmelCase )
self.assertDictEqual(dset['train'][0] , get_test_nested_examples()[0][1] )
self.assertDictEqual(
dset['train'][expected_num_examples - 1] , get_test_nested_examples()[expected_num_examples - 1][1] )
self.assertTrue(
os.path.exists(os.path.join(_UpperCAmelCase , builder.name , 'default' , '0.0.0' , 'dataset_info.json' ) ) )
del dset
| 23 | 0 |
def __SCREAMING_SNAKE_CASE ( a__ : list ,a__ : list ,a__ : int ,a__ : int ,a__ : int ) -> int:
if index == number_of_items:
return 0
__A : Optional[int] = 0
__A : List[Any] = 0
__A : int = knapsack(a__ ,a__ ,a__ ,a__ ,index + 1 )
if weights[index] <= max_weight:
__A : Union[str, Any] = values[index] + knapsack(
a__ ,a__ ,a__ ,max_weight - weights[index] ,index + 1 )
return max(a__ ,a__ )
if __name__ == "__main__":
import doctest
doctest.testmod()
| 17 |
import argparse
import torch
from ...utils import logging
from . import AlbertConfig, AlbertForPreTraining, load_tf_weights_in_albert
logging.set_verbosity_info()
def _snake_case (__lowercase , __lowercase , __lowercase):
# Initialise PyTorch model
UpperCamelCase_ = AlbertConfig.from_json_file(__lowercase)
print(f"""Building PyTorch model from configuration: {config}""")
UpperCamelCase_ = AlbertForPreTraining(__lowercase)
# Load weights from tf checkpoint
load_tf_weights_in_albert(__lowercase , __lowercase , __lowercase)
# Save pytorch-model
print(f"""Save PyTorch model to {pytorch_dump_path}""")
torch.save(model.state_dict() , __lowercase)
if __name__ == "__main__":
snake_case__ : Tuple = argparse.ArgumentParser()
# Required parameters
parser.add_argument(
"""--tf_checkpoint_path""", default=None, type=str, required=True, help="""Path to the TensorFlow checkpoint path."""
)
parser.add_argument(
"""--albert_config_file""",
default=None,
type=str,
required=True,
help=(
"""The config json file corresponding to the pre-trained ALBERT model. \n"""
"""This specifies the model architecture."""
),
)
parser.add_argument(
"""--pytorch_dump_path""", default=None, type=str, required=True, help="""Path to the output PyTorch model."""
)
snake_case__ : str = parser.parse_args()
convert_tf_checkpoint_to_pytorch(args.tf_checkpoint_path, args.albert_config_file, args.pytorch_dump_path)
| 23 | 0 |
'''simple docstring'''
from typing import TYPE_CHECKING
from ...utils import (
OptionalDependencyNotAvailable,
_LazyModule,
is_flax_available,
is_tf_available,
is_torch_available,
is_vision_available,
)
_SCREAMING_SNAKE_CASE = {"configuration_vit": ["VIT_PRETRAINED_CONFIG_ARCHIVE_MAP", "ViTConfig", "ViTOnnxConfig"]}
try:
if not is_vision_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
_SCREAMING_SNAKE_CASE = ["ViTFeatureExtractor"]
_SCREAMING_SNAKE_CASE = ["ViTImageProcessor"]
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
_SCREAMING_SNAKE_CASE = [
"VIT_PRETRAINED_MODEL_ARCHIVE_LIST",
"ViTForImageClassification",
"ViTForMaskedImageModeling",
"ViTModel",
"ViTPreTrainedModel",
]
try:
if not is_tf_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
_SCREAMING_SNAKE_CASE = [
"TFViTForImageClassification",
"TFViTModel",
"TFViTPreTrainedModel",
]
try:
if not is_flax_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
_SCREAMING_SNAKE_CASE = [
"FlaxViTForImageClassification",
"FlaxViTModel",
"FlaxViTPreTrainedModel",
]
if TYPE_CHECKING:
from .configuration_vit import VIT_PRETRAINED_CONFIG_ARCHIVE_MAP, ViTConfig, ViTOnnxConfig
try:
if not is_vision_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .feature_extraction_vit import ViTFeatureExtractor
from .image_processing_vit import ViTImageProcessor
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_vit import (
VIT_PRETRAINED_MODEL_ARCHIVE_LIST,
ViTForImageClassification,
ViTForMaskedImageModeling,
ViTModel,
ViTPreTrainedModel,
)
try:
if not is_tf_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_tf_vit import TFViTForImageClassification, TFViTModel, TFViTPreTrainedModel
try:
if not is_flax_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_flax_vit import FlaxViTForImageClassification, FlaxViTModel, FlaxViTPreTrainedModel
else:
import sys
_SCREAMING_SNAKE_CASE = _LazyModule(__name__, globals()["__file__"], _import_structure, module_spec=__spec__)
| 18 |
from transformers import BertTokenizer, EncoderDecoderModel, SeqaSeqTrainer, SeqaSeqTrainingArguments
from transformers.testing_utils import TestCasePlus, require_torch, slow
from transformers.utils import is_datasets_available
if is_datasets_available():
import datasets
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
@slow
@require_torch
def _UpperCAmelCase ( self ) -> List[Any]:
UpperCamelCase_ = EncoderDecoderModel.from_encoder_decoder_pretrained('prajjwal1/bert-tiny' , 'prajjwal1/bert-tiny' )
UpperCamelCase_ = BertTokenizer.from_pretrained('bert-base-uncased' )
UpperCamelCase_ = bertabert.config.encoder.vocab_size
UpperCamelCase_ = tokenizer.sep_token_id
UpperCamelCase_ = tokenizer.cls_token_id
UpperCamelCase_ = 128
UpperCamelCase_ = datasets.load_dataset('cnn_dailymail' , '3.0.0' , split='train[:1%]' )
UpperCamelCase_ = datasets.load_dataset('cnn_dailymail' , '3.0.0' , split='validation[:1%]' )
UpperCamelCase_ = train_dataset.select(range(32 ) )
UpperCamelCase_ = val_dataset.select(range(16 ) )
UpperCamelCase_ = 4
def _map_to_encoder_decoder_inputs(_UpperCAmelCase ):
# Tokenizer will automatically set [BOS] <text> [EOS]
UpperCamelCase_ = tokenizer(batch['article'] , padding='max_length' , truncation=_UpperCAmelCase , max_length=512 )
UpperCamelCase_ = tokenizer(batch['highlights'] , padding='max_length' , truncation=_UpperCAmelCase , max_length=128 )
UpperCamelCase_ = inputs.input_ids
UpperCamelCase_ = inputs.attention_mask
UpperCamelCase_ = outputs.input_ids
UpperCamelCase_ = outputs.input_ids.copy()
UpperCamelCase_ = [
[-100 if token == tokenizer.pad_token_id else token for token in labels] for labels in batch['labels']
]
UpperCamelCase_ = outputs.attention_mask
assert all(len(_UpperCAmelCase ) == 512 for x in inputs.input_ids )
assert all(len(_UpperCAmelCase ) == 128 for x in outputs.input_ids )
return batch
def _compute_metrics(_UpperCAmelCase ):
UpperCamelCase_ = pred.label_ids
UpperCamelCase_ = pred.predictions
# all unnecessary tokens are removed
UpperCamelCase_ = tokenizer.batch_decode(_UpperCAmelCase , skip_special_tokens=_UpperCAmelCase )
UpperCamelCase_ = tokenizer.batch_decode(_UpperCAmelCase , skip_special_tokens=_UpperCAmelCase )
UpperCamelCase_ = sum([int(pred_str[i] == label_str[i] ) for i in range(len(_UpperCAmelCase ) )] ) / len(_UpperCAmelCase )
return {"accuracy": accuracy}
# map train dataset
UpperCamelCase_ = train_dataset.map(
_map_to_encoder_decoder_inputs , batched=_UpperCAmelCase , batch_size=_UpperCAmelCase , remove_columns=['article', 'highlights'] , )
train_dataset.set_format(
type='torch' , columns=['input_ids', 'attention_mask', 'decoder_input_ids', 'decoder_attention_mask', 'labels'] , )
# same for validation dataset
UpperCamelCase_ = val_dataset.map(
_map_to_encoder_decoder_inputs , batched=_UpperCAmelCase , batch_size=_UpperCAmelCase , remove_columns=['article', 'highlights'] , )
val_dataset.set_format(
type='torch' , columns=['input_ids', 'attention_mask', 'decoder_input_ids', 'decoder_attention_mask', 'labels'] , )
UpperCamelCase_ = self.get_auto_remove_tmp_dir()
UpperCamelCase_ = SeqaSeqTrainingArguments(
output_dir=_UpperCAmelCase , per_device_train_batch_size=_UpperCAmelCase , per_device_eval_batch_size=_UpperCAmelCase , predict_with_generate=_UpperCAmelCase , evaluation_strategy='steps' , do_train=_UpperCAmelCase , do_eval=_UpperCAmelCase , warmup_steps=0 , eval_steps=2 , logging_steps=2 , )
# instantiate trainer
UpperCamelCase_ = SeqaSeqTrainer(
model=_UpperCAmelCase , args=_UpperCAmelCase , compute_metrics=_compute_metrics , train_dataset=_UpperCAmelCase , eval_dataset=_UpperCAmelCase , tokenizer=_UpperCAmelCase , )
# start training
trainer.train()
| 23 | 0 |
"""simple docstring"""
import math
def lowerCamelCase__ ( __snake_case ) -> bool:
"""simple docstring"""
assert isinstance(__snake_case, __snake_case ) and (
number >= 0
), "'number' must been an int and positive"
if 1 < number < 4:
# 2 and 3 are primes
return True
elif number < 2 or not number % 2:
# Negatives, 0, 1 and all even numbers are not primes
return False
_UpperCamelCase = range(3, int(math.sqrt(__snake_case ) + 1 ), 2 )
return not any(not number % i for i in odd_numbers )
def lowerCamelCase__ ( __snake_case, __snake_case=1, **__snake_case ) -> int:
"""simple docstring"""
_UpperCamelCase = factor * value
_UpperCamelCase = value
while not is_prime(__snake_case ):
value += 1 if not ("desc" in kwargs and kwargs["desc"] is True) else -1
if value == first_value_val:
return next_prime(value + 1, **__snake_case )
return value
| 19 |
import argparse
import os
import evaluate
import torch
from datasets import load_dataset
from torch.optim import AdamW
from torch.utils.data import DataLoader
from transformers import AutoModelForSequenceClassification, AutoTokenizer, get_linear_schedule_with_warmup, set_seed
from accelerate import Accelerator, DistributedType
########################################################################
# This is a fully working simple example to use Accelerate
# and perform gradient accumulation
#
# This example trains a Bert base model on GLUE MRPC
# in any of the following settings (with the same script):
# - single CPU or single GPU
# - multi GPUS (using PyTorch distributed mode)
# - (multi) TPUs
# - fp16 (mixed-precision) or fp32 (normal precision)
#
# To run it in each of these various modes, follow the instructions
# in the readme for examples:
# https://github.com/huggingface/accelerate/tree/main/examples
#
########################################################################
snake_case__ : Dict = 1_6
snake_case__ : List[str] = 3_2
def _snake_case (__lowercase , __lowercase = 16):
UpperCamelCase_ = AutoTokenizer.from_pretrained('bert-base-cased')
UpperCamelCase_ = load_dataset('glue' , 'mrpc')
def tokenize_function(__lowercase):
# max_length=None => use the model max length (it's actually the default)
UpperCamelCase_ = tokenizer(examples['sentence1'] , examples['sentence2'] , truncation=__lowercase , max_length=__lowercase)
return outputs
# Apply the method we just defined to all the examples in all the splits of the dataset
# starting with the main process first:
with accelerator.main_process_first():
UpperCamelCase_ = datasets.map(
__lowercase , batched=__lowercase , remove_columns=['idx', 'sentence1', 'sentence2'] , )
# We also rename the 'label' column to 'labels' which is the expected name for labels by the models of the
# transformers library
UpperCamelCase_ = tokenized_datasets.rename_column('label' , 'labels')
def collate_fn(__lowercase):
# On TPU it's best to pad everything to the same length or training will be very slow.
UpperCamelCase_ = 128 if accelerator.distributed_type == DistributedType.TPU else None
# When using mixed precision we want round multiples of 8/16
if accelerator.mixed_precision == "fp8":
UpperCamelCase_ = 16
elif accelerator.mixed_precision != "no":
UpperCamelCase_ = 8
else:
UpperCamelCase_ = None
return tokenizer.pad(
__lowercase , padding='longest' , max_length=__lowercase , pad_to_multiple_of=__lowercase , return_tensors='pt' , )
# Instantiate dataloaders.
UpperCamelCase_ = DataLoader(
tokenized_datasets['train'] , shuffle=__lowercase , collate_fn=__lowercase , batch_size=__lowercase)
UpperCamelCase_ = DataLoader(
tokenized_datasets['validation'] , shuffle=__lowercase , collate_fn=__lowercase , batch_size=__lowercase)
return train_dataloader, eval_dataloader
# For testing only
if os.environ.get("""TESTING_MOCKED_DATALOADERS""", None) == "1":
from accelerate.test_utils.training import mocked_dataloaders
snake_case__ : List[str] = mocked_dataloaders # noqa: F811
def _snake_case (__lowercase , __lowercase):
# For testing only
if os.environ.get('TESTING_MOCKED_DATALOADERS' , __lowercase) == "1":
UpperCamelCase_ = 2
# New Code #
UpperCamelCase_ = int(args.gradient_accumulation_steps)
# Initialize accelerator
UpperCamelCase_ = Accelerator(
cpu=args.cpu , mixed_precision=args.mixed_precision , gradient_accumulation_steps=__lowercase)
if accelerator.distributed_type == DistributedType.TPU and gradient_accumulation_steps > 1:
raise NotImplementedError(
'Gradient accumulation on TPUs is currently not supported. Pass `gradient_accumulation_steps=1`')
# Sample hyper-parameters for learning rate, batch size, seed and a few other HPs
UpperCamelCase_ = config['lr']
UpperCamelCase_ = int(config['num_epochs'])
UpperCamelCase_ = int(config['seed'])
UpperCamelCase_ = int(config['batch_size'])
UpperCamelCase_ = evaluate.load('glue' , 'mrpc')
set_seed(__lowercase)
UpperCamelCase_ , UpperCamelCase_ = get_dataloaders(__lowercase , __lowercase)
# Instantiate the model (we build the model here so that the seed also control new weights initialization)
UpperCamelCase_ = AutoModelForSequenceClassification.from_pretrained('bert-base-cased' , return_dict=__lowercase)
# We could avoid this line since the accelerator is set with `device_placement=True` (default value).
# Note that if you are placing tensors on devices manually, this line absolutely needs to be before the optimizer
# creation otherwise training will not work on TPU (`accelerate` will kindly throw an error to make us aware of that).
UpperCamelCase_ = model.to(accelerator.device)
# Instantiate optimizer
UpperCamelCase_ = AdamW(params=model.parameters() , lr=__lowercase)
# Instantiate scheduler
UpperCamelCase_ = get_linear_schedule_with_warmup(
optimizer=__lowercase , num_warmup_steps=100 , num_training_steps=(len(__lowercase) * num_epochs) , )
# Prepare everything
# There is no specific order to remember, we just need to unpack the objects in the same order we gave them to the
# prepare method.
UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ = accelerator.prepare(
__lowercase , __lowercase , __lowercase , __lowercase , __lowercase)
# Now we train the model
for epoch in range(__lowercase):
model.train()
for step, batch in enumerate(__lowercase):
# We could avoid this line since we set the accelerator with `device_placement=True`.
batch.to(accelerator.device)
# New code #
# We use the new `accumulate` context manager to perform gradient accumulation
# We also currently do not support TPUs nor advise it as bugs were found on the XLA side when running our tests.
with accelerator.accumulate(__lowercase):
UpperCamelCase_ = model(**__lowercase)
UpperCamelCase_ = output.loss
accelerator.backward(__lowercase)
optimizer.step()
lr_scheduler.step()
optimizer.zero_grad()
model.eval()
for step, batch in enumerate(__lowercase):
# We could avoid this line since we set the accelerator with `device_placement=True`.
batch.to(accelerator.device)
with torch.no_grad():
UpperCamelCase_ = model(**__lowercase)
UpperCamelCase_ = outputs.logits.argmax(dim=-1)
UpperCamelCase_ , UpperCamelCase_ = accelerator.gather_for_metrics((predictions, batch['labels']))
metric.add_batch(
predictions=__lowercase , references=__lowercase , )
UpperCamelCase_ = metric.compute()
# Use accelerator.print to print only on the main process.
accelerator.print(f"""epoch {epoch}:""" , __lowercase)
def _snake_case ():
UpperCamelCase_ = argparse.ArgumentParser(description='Simple example of training script.')
parser.add_argument(
'--mixed_precision' , type=__lowercase , default=__lowercase , choices=['no', 'fp16', 'bf16', 'fp8'] , help='Whether to use mixed precision. Choose'
'between fp16 and bf16 (bfloat16). Bf16 requires PyTorch >= 1.10.'
'and an Nvidia Ampere GPU.' , )
# New Code #
parser.add_argument(
'--gradient_accumulation_steps' , type=__lowercase , default=1 , help='The number of minibatches to be ran before gradients are accumulated.' , )
parser.add_argument('--cpu' , action='store_true' , help='If passed, will train on the CPU.')
UpperCamelCase_ = parser.parse_args()
UpperCamelCase_ = {'lr': 2e-5, 'num_epochs': 3, 'seed': 42, 'batch_size': 16}
training_function(__lowercase , __lowercase)
if __name__ == "__main__":
main()
| 23 | 0 |
from collections import OrderedDict
from typing import Mapping
from packaging import version
from ...configuration_utils import PretrainedConfig
from ...onnx import OnnxConfig
from ...utils import logging
_lowerCAmelCase: Optional[int] = logging.get_logger(__name__)
_lowerCAmelCase: Any = {
'microsoft/beit-base-patch16-224-pt22k': (
'https://huggingface.co/microsoft/beit-base-patch16-224-pt22k/resolve/main/config.json'
),
# See all BEiT models at https://huggingface.co/models?filter=beit
}
class lowercase_ (lowercase__ ):
snake_case ='beit'
def __init__( self , lowercase_=8192 , lowercase_=768 , lowercase_=12 , lowercase_=12 , lowercase_=3072 , lowercase_="gelu" , lowercase_=0.0 , lowercase_=0.0 , lowercase_=0.02 , lowercase_=1e-12 , lowercase_=224 , lowercase_=16 , lowercase_=3 , lowercase_=False , lowercase_=False , lowercase_=False , lowercase_=False , lowercase_=0.1 , lowercase_=0.1 , lowercase_=True , lowercase_=[3, 5, 7, 11] , lowercase_=[1, 2, 3, 6] , lowercase_=True , lowercase_=0.4 , lowercase_=256 , lowercase_=1 , lowercase_=False , lowercase_=255 , **lowercase_ , ) -> Optional[int]:
super().__init__(**lowercase_)
a__ =vocab_size
a__ =hidden_size
a__ =num_hidden_layers
a__ =num_attention_heads
a__ =intermediate_size
a__ =hidden_act
a__ =hidden_dropout_prob
a__ =attention_probs_dropout_prob
a__ =initializer_range
a__ =layer_norm_eps
a__ =image_size
a__ =patch_size
a__ =num_channels
a__ =use_mask_token
a__ =use_absolute_position_embeddings
a__ =use_relative_position_bias
a__ =use_shared_relative_position_bias
a__ =layer_scale_init_value
a__ =drop_path_rate
a__ =use_mean_pooling
# decode head attributes (semantic segmentation)
a__ =out_indices
a__ =pool_scales
# auxiliary head attributes (semantic segmentation)
a__ =use_auxiliary_head
a__ =auxiliary_loss_weight
a__ =auxiliary_channels
a__ =auxiliary_num_convs
a__ =auxiliary_concat_input
a__ =semantic_loss_ignore_index
class lowercase_ (lowercase__ ):
snake_case =version.parse('1.11' )
@property
def __UpperCamelCase ( self) -> Mapping[str, Mapping[int, str]]:
return OrderedDict(
[
('pixel_values', {0: 'batch', 1: 'num_channels', 2: 'height', 3: 'width'}),
])
@property
def __UpperCamelCase ( self) -> float:
return 1e-4
| 20 |
import inspect
import unittest
import numpy as np
from tests.test_modeling_common import floats_tensor
from transformers import MaskaFormerConfig, is_torch_available, is_vision_available
from transformers.testing_utils import require_torch, require_torch_multi_gpu, require_vision, slow, torch_device
from transformers.utils import cached_property
from ...test_configuration_common import ConfigTester
from ...test_modeling_common import ModelTesterMixin
from ...test_pipeline_mixin import PipelineTesterMixin
if is_torch_available():
import torch
from transformers import MaskaFormerForUniversalSegmentation, MaskaFormerModel
if is_vision_available():
from transformers import MaskaFormerImageProcessor
if is_vision_available():
from PIL import Image
class _a :
"""simple docstring"""
def __init__( self , _UpperCAmelCase , _UpperCAmelCase=2 , _UpperCAmelCase=True , _UpperCAmelCase=False , _UpperCAmelCase=10 , _UpperCAmelCase=3 , _UpperCAmelCase=32 * 8 , _UpperCAmelCase=32 * 8 , _UpperCAmelCase=4 , _UpperCAmelCase=64 , ) -> List[Any]:
UpperCamelCase_ = parent
UpperCamelCase_ = batch_size
UpperCamelCase_ = is_training
UpperCamelCase_ = use_auxiliary_loss
UpperCamelCase_ = num_queries
UpperCamelCase_ = num_channels
UpperCamelCase_ = min_size
UpperCamelCase_ = max_size
UpperCamelCase_ = num_labels
UpperCamelCase_ = hidden_dim
UpperCamelCase_ = hidden_dim
def _UpperCAmelCase ( self ) -> List[str]:
UpperCamelCase_ = floats_tensor([self.batch_size, self.num_channels, self.min_size, self.max_size] ).to(
_UpperCAmelCase )
UpperCamelCase_ = torch.ones([self.batch_size, self.min_size, self.max_size] , device=_UpperCAmelCase )
UpperCamelCase_ = (
torch.rand([self.batch_size, self.num_labels, self.min_size, self.max_size] , device=_UpperCAmelCase ) > 0.5
).float()
UpperCamelCase_ = (torch.rand((self.batch_size, self.num_labels) , device=_UpperCAmelCase ) > 0.5).long()
UpperCamelCase_ = self.get_config()
return config, pixel_values, pixel_mask, mask_labels, class_labels
def _UpperCAmelCase ( self ) -> Optional[int]:
UpperCamelCase_ = MaskaFormerConfig(
hidden_size=self.hidden_dim , )
UpperCamelCase_ = self.num_queries
UpperCamelCase_ = self.num_labels
UpperCamelCase_ = [1, 1, 1, 1]
UpperCamelCase_ = self.num_channels
UpperCamelCase_ = 64
UpperCamelCase_ = 128
UpperCamelCase_ = self.hidden_dim
UpperCamelCase_ = self.hidden_dim
UpperCamelCase_ = self.hidden_dim
return config
def _UpperCAmelCase ( self ) -> Dict:
UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ = self.prepare_config_and_inputs()
UpperCamelCase_ = {'pixel_values': pixel_values, 'pixel_mask': pixel_mask}
return config, inputs_dict
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> List[Any]:
UpperCamelCase_ = output.encoder_hidden_states
UpperCamelCase_ = output.pixel_decoder_hidden_states
UpperCamelCase_ = output.transformer_decoder_hidden_states
self.parent.assertTrue(len(_UpperCAmelCase ) , len(config.backbone_config.depths ) )
self.parent.assertTrue(len(_UpperCAmelCase ) , len(config.backbone_config.depths ) )
self.parent.assertTrue(len(_UpperCAmelCase ) , config.decoder_layers )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase=False ) -> Any:
with torch.no_grad():
UpperCamelCase_ = MaskaFormerModel(config=_UpperCAmelCase )
model.to(_UpperCAmelCase )
model.eval()
UpperCamelCase_ = model(pixel_values=_UpperCAmelCase , pixel_mask=_UpperCAmelCase )
UpperCamelCase_ = model(_UpperCAmelCase , output_hidden_states=_UpperCAmelCase )
self.parent.assertEqual(
output.transformer_decoder_last_hidden_state.shape , (self.batch_size, self.num_queries, self.hidden_dim) , )
# let's ensure the other two hidden state exists
self.parent.assertTrue(output.pixel_decoder_last_hidden_state is not None )
self.parent.assertTrue(output.encoder_last_hidden_state is not None )
if output_hidden_states:
self.check_output_hidden_state(_UpperCAmelCase , _UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase ) -> Union[str, Any]:
UpperCamelCase_ = MaskaFormerForUniversalSegmentation(config=_UpperCAmelCase )
model.to(_UpperCAmelCase )
model.eval()
def comm_check_on_output(_UpperCAmelCase ):
# let's still check that all the required stuff is there
self.parent.assertTrue(result.transformer_decoder_last_hidden_state is not None )
self.parent.assertTrue(result.pixel_decoder_last_hidden_state is not None )
self.parent.assertTrue(result.encoder_last_hidden_state is not None )
# okay, now we need to check the logits shape
# due to the encoder compression, masks have a //4 spatial size
self.parent.assertEqual(
result.masks_queries_logits.shape , (self.batch_size, self.num_queries, self.min_size // 4, self.max_size // 4) , )
# + 1 for null class
self.parent.assertEqual(
result.class_queries_logits.shape , (self.batch_size, self.num_queries, self.num_labels + 1) )
with torch.no_grad():
UpperCamelCase_ = model(pixel_values=_UpperCAmelCase , pixel_mask=_UpperCAmelCase )
UpperCamelCase_ = model(_UpperCAmelCase )
comm_check_on_output(_UpperCAmelCase )
UpperCamelCase_ = model(
pixel_values=_UpperCAmelCase , pixel_mask=_UpperCAmelCase , mask_labels=_UpperCAmelCase , class_labels=_UpperCAmelCase )
comm_check_on_output(_UpperCAmelCase )
self.parent.assertTrue(result.loss is not None )
self.parent.assertEqual(result.loss.shape , torch.Size([1] ) )
@require_torch
class _a ( UpperCAmelCase__ , UpperCAmelCase__ , unittest.TestCase ):
"""simple docstring"""
A_ = (MaskaFormerModel, MaskaFormerForUniversalSegmentation) if is_torch_available() else ()
A_ = {"""feature-extraction""": MaskaFormerModel} if is_torch_available() else {}
A_ = False
A_ = False
A_ = False
A_ = False
def _UpperCAmelCase ( self ) -> Optional[Any]:
UpperCamelCase_ = MaskaFormerModelTester(self )
UpperCamelCase_ = ConfigTester(self , config_class=_UpperCAmelCase , has_text_modality=_UpperCAmelCase )
def _UpperCAmelCase ( self ) -> Union[str, Any]:
self.config_tester.run_common_tests()
def _UpperCAmelCase ( self ) -> Union[str, Any]:
UpperCamelCase_ , UpperCamelCase_ = self.model_tester.prepare_config_and_inputs_for_common()
self.model_tester.create_and_check_maskaformer_model(_UpperCAmelCase , **_UpperCAmelCase , output_hidden_states=_UpperCAmelCase )
def _UpperCAmelCase ( self ) -> List[Any]:
UpperCamelCase_ = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_maskaformer_instance_segmentation_head_model(*_UpperCAmelCase )
@unittest.skip(reason='Mask2Former does not use inputs_embeds' )
def _UpperCAmelCase ( self ) -> Any:
pass
@unittest.skip(reason='Mask2Former does not have a get_input_embeddings method' )
def _UpperCAmelCase ( self ) -> Optional[int]:
pass
@unittest.skip(reason='Mask2Former is not a generative model' )
def _UpperCAmelCase ( self ) -> Any:
pass
@unittest.skip(reason='Mask2Former does not use token embeddings' )
def _UpperCAmelCase ( self ) -> Optional[Any]:
pass
@require_torch_multi_gpu
@unittest.skip(
reason='Mask2Former has some layers using `add_module` which doesn\'t work well with `nn.DataParallel`' )
def _UpperCAmelCase ( self ) -> int:
pass
@unittest.skip('Will be fixed soon by reducing the size of the model used for common tests.' )
def _UpperCAmelCase ( self ) -> str:
pass
def _UpperCAmelCase ( self ) -> Optional[int]:
UpperCamelCase_ , UpperCamelCase_ = self.model_tester.prepare_config_and_inputs_for_common()
for model_class in self.all_model_classes:
UpperCamelCase_ = model_class(_UpperCAmelCase )
UpperCamelCase_ = inspect.signature(model.forward )
# signature.parameters is an OrderedDict => so arg_names order is deterministic
UpperCamelCase_ = [*signature.parameters.keys()]
UpperCamelCase_ = ['pixel_values']
self.assertListEqual(arg_names[:1] , _UpperCAmelCase )
@slow
def _UpperCAmelCase ( self ) -> Tuple:
for model_name in ["facebook/mask2former-swin-small-coco-instance"]:
UpperCamelCase_ = MaskaFormerModel.from_pretrained(_UpperCAmelCase )
self.assertIsNotNone(_UpperCAmelCase )
def _UpperCAmelCase ( self ) -> Dict:
UpperCamelCase_ = (self.model_tester.min_size,) * 2
UpperCamelCase_ = {
'pixel_values': torch.randn((2, 3, *size) , device=_UpperCAmelCase ),
'mask_labels': torch.randn((2, 10, *size) , device=_UpperCAmelCase ),
'class_labels': torch.zeros(2 , 10 , device=_UpperCAmelCase ).long(),
}
UpperCamelCase_ = self.model_tester.get_config()
UpperCamelCase_ = MaskaFormerForUniversalSegmentation(_UpperCAmelCase ).to(_UpperCAmelCase )
UpperCamelCase_ = model(**_UpperCAmelCase )
self.assertTrue(outputs.loss is not None )
def _UpperCAmelCase ( self ) -> str:
UpperCamelCase_ , UpperCamelCase_ = self.model_tester.prepare_config_and_inputs_for_common()
self.model_tester.create_and_check_maskaformer_model(_UpperCAmelCase , **_UpperCAmelCase , output_hidden_states=_UpperCAmelCase )
def _UpperCAmelCase ( self ) -> Optional[int]:
UpperCamelCase_ , UpperCamelCase_ = self.model_tester.prepare_config_and_inputs_for_common()
for model_class in self.all_model_classes:
UpperCamelCase_ = model_class(_UpperCAmelCase ).to(_UpperCAmelCase )
UpperCamelCase_ = model(**_UpperCAmelCase , output_attentions=_UpperCAmelCase )
self.assertTrue(outputs.attentions is not None )
def _UpperCAmelCase ( self ) -> List[Any]:
if not self.model_tester.is_training:
return
UpperCamelCase_ = self.all_model_classes[1]
UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ = self.model_tester.prepare_config_and_inputs()
UpperCamelCase_ = model_class(_UpperCAmelCase )
model.to(_UpperCAmelCase )
model.train()
UpperCamelCase_ = model(_UpperCAmelCase , mask_labels=_UpperCAmelCase , class_labels=_UpperCAmelCase ).loss
loss.backward()
def _UpperCAmelCase ( self ) -> int:
UpperCamelCase_ = self.all_model_classes[1]
UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ = self.model_tester.prepare_config_and_inputs()
UpperCamelCase_ = True
UpperCamelCase_ = True
UpperCamelCase_ = model_class(_UpperCAmelCase ).to(_UpperCAmelCase )
model.train()
UpperCamelCase_ = model(_UpperCAmelCase , mask_labels=_UpperCAmelCase , class_labels=_UpperCAmelCase )
UpperCamelCase_ = outputs.encoder_hidden_states[0]
encoder_hidden_states.retain_grad()
UpperCamelCase_ = outputs.pixel_decoder_hidden_states[0]
pixel_decoder_hidden_states.retain_grad()
UpperCamelCase_ = outputs.transformer_decoder_hidden_states[0]
transformer_decoder_hidden_states.retain_grad()
UpperCamelCase_ = outputs.attentions[0]
attentions.retain_grad()
outputs.loss.backward(retain_graph=_UpperCAmelCase )
self.assertIsNotNone(encoder_hidden_states.grad )
self.assertIsNotNone(pixel_decoder_hidden_states.grad )
self.assertIsNotNone(transformer_decoder_hidden_states.grad )
self.assertIsNotNone(attentions.grad )
snake_case__ : List[Any] = 1E-4
def _snake_case ():
UpperCamelCase_ = Image.open('./tests/fixtures/tests_samples/COCO/000000039769.png')
return image
@require_vision
@slow
class _a ( unittest.TestCase ):
"""simple docstring"""
@cached_property
def _UpperCAmelCase ( self ) -> Optional[int]:
return "facebook/mask2former-swin-small-coco-instance"
@cached_property
def _UpperCAmelCase ( self ) -> List[str]:
return MaskaFormerImageProcessor.from_pretrained(self.model_checkpoints ) if is_vision_available() else None
def _UpperCAmelCase ( self ) -> str:
UpperCamelCase_ = MaskaFormerModel.from_pretrained(self.model_checkpoints ).to(_UpperCAmelCase )
UpperCamelCase_ = self.default_image_processor
UpperCamelCase_ = prepare_img()
UpperCamelCase_ = image_processor(_UpperCAmelCase , return_tensors='pt' ).to(_UpperCAmelCase )
UpperCamelCase_ = inputs['pixel_values'].shape
# check size is divisible by 32
self.assertTrue((inputs_shape[-1] % 32) == 0 and (inputs_shape[-2] % 32) == 0 )
# check size
self.assertEqual(_UpperCAmelCase , (1, 3, 384, 384) )
with torch.no_grad():
UpperCamelCase_ = model(**_UpperCAmelCase )
UpperCamelCase_ = torch.tensor(
[[-0.2_7_9_0, -1.0_7_1_7, -1.1_6_6_8], [-0.5_1_2_8, -0.3_1_2_8, -0.4_9_8_7], [-0.5_8_3_2, 0.1_9_7_1, -0.0_1_9_7]] ).to(_UpperCAmelCase )
self.assertTrue(
torch.allclose(
outputs.encoder_last_hidden_state[0, 0, :3, :3] , _UpperCAmelCase , atol=_UpperCAmelCase ) )
UpperCamelCase_ = torch.tensor(
[[0.8_9_7_3, 1.1_8_4_7, 1.1_7_7_6], [1.1_9_3_4, 1.5_0_4_0, 1.5_1_2_8], [1.1_1_5_3, 1.4_4_8_6, 1.4_9_5_1]] ).to(_UpperCAmelCase )
self.assertTrue(
torch.allclose(
outputs.pixel_decoder_last_hidden_state[0, 0, :3, :3] , _UpperCAmelCase , atol=_UpperCAmelCase ) )
UpperCamelCase_ = torch.tensor(
[[2.1_1_5_2, 1.7_0_0_0, -0.8_6_0_3], [1.5_8_0_8, 1.8_0_0_4, -0.9_3_5_3], [1.6_0_4_3, 1.7_4_9_5, -0.5_9_9_9]] ).to(_UpperCAmelCase )
self.assertTrue(
torch.allclose(
outputs.transformer_decoder_last_hidden_state[0, :3, :3] , _UpperCAmelCase , atol=_UpperCAmelCase ) )
def _UpperCAmelCase ( self ) -> Optional[Any]:
UpperCamelCase_ = MaskaFormerForUniversalSegmentation.from_pretrained(self.model_checkpoints ).to(_UpperCAmelCase ).eval()
UpperCamelCase_ = self.default_image_processor
UpperCamelCase_ = prepare_img()
UpperCamelCase_ = image_processor(_UpperCAmelCase , return_tensors='pt' ).to(_UpperCAmelCase )
UpperCamelCase_ = inputs['pixel_values'].shape
# check size is divisible by 32
self.assertTrue((inputs_shape[-1] % 32) == 0 and (inputs_shape[-2] % 32) == 0 )
# check size
self.assertEqual(_UpperCAmelCase , (1, 3, 384, 384) )
with torch.no_grad():
UpperCamelCase_ = model(**_UpperCAmelCase )
# masks_queries_logits
UpperCamelCase_ = outputs.masks_queries_logits
self.assertEqual(
masks_queries_logits.shape , (1, model.config.num_queries, inputs_shape[-2] // 4, inputs_shape[-1] // 4) )
UpperCamelCase_ = [
[-8.7_8_3_9, -9.0_0_5_6, -8.8_1_2_1],
[-7.4_1_0_4, -7.0_3_1_3, -6.5_4_0_1],
[-6.6_1_0_5, -6.3_4_2_7, -6.4_6_7_5],
]
UpperCamelCase_ = torch.tensor(_UpperCAmelCase ).to(_UpperCAmelCase )
self.assertTrue(torch.allclose(masks_queries_logits[0, 0, :3, :3] , _UpperCAmelCase , atol=_UpperCAmelCase ) )
# class_queries_logits
UpperCamelCase_ = outputs.class_queries_logits
self.assertEqual(class_queries_logits.shape , (1, model.config.num_queries, model.config.num_labels + 1) )
UpperCamelCase_ = torch.tensor(
[
[1.8_3_2_4, -8.0_8_3_5, -4.1_9_2_2],
[0.8_4_5_0, -9.0_0_5_0, -3.6_0_5_3],
[0.3_0_4_5, -7.7_2_9_3, -3.0_2_7_5],
] ).to(_UpperCAmelCase )
self.assertTrue(torch.allclose(outputs.class_queries_logits[0, :3, :3] , _UpperCAmelCase , atol=_UpperCAmelCase ) )
def _UpperCAmelCase ( self ) -> Dict:
UpperCamelCase_ = MaskaFormerForUniversalSegmentation.from_pretrained(self.model_checkpoints ).to(_UpperCAmelCase ).eval()
UpperCamelCase_ = self.default_image_processor
UpperCamelCase_ = image_processor(
[np.zeros((3, 800, 1333) ), np.zeros((3, 800, 1333) )] , segmentation_maps=[np.zeros((384, 384) ).astype(np.floataa ), np.zeros((384, 384) ).astype(np.floataa )] , return_tensors='pt' , )
UpperCamelCase_ = inputs['pixel_values'].to(_UpperCAmelCase )
UpperCamelCase_ = [el.to(_UpperCAmelCase ) for el in inputs['mask_labels']]
UpperCamelCase_ = [el.to(_UpperCAmelCase ) for el in inputs['class_labels']]
with torch.no_grad():
UpperCamelCase_ = model(**_UpperCAmelCase )
self.assertTrue(outputs.loss is not None )
| 23 | 0 |
import functools
import gc
import inspect
import torch
from .imports import is_npu_available, is_xpu_available
def lowerCAmelCase_ ( *lowerCamelCase ):
if not isinstance(lowerCamelCase , lowerCamelCase ):
__magic_name__ : Union[str, Any] =list(lowerCamelCase )
for i in range(len(lowerCamelCase ) ):
__magic_name__ : Dict =None
gc.collect()
if is_xpu_available():
torch.xpu.empty_cache()
elif is_npu_available():
torch.npu.empty_cache()
else:
torch.cuda.empty_cache()
return objects
def lowerCAmelCase_ ( lowerCamelCase ):
__magic_name__ : List[str] =[
"""CUDA out of memory.""", # CUDA OOM
"""cuDNN error: CUDNN_STATUS_NOT_SUPPORTED.""", # CUDNN SNAFU
"""DefaultCPUAllocator: can't allocate memory""", # CPU OOM
]
if isinstance(lowerCamelCase , lowerCamelCase ) and len(exception.args ) == 1:
return any(err in exception.args[0] for err in _statements )
return False
def lowerCAmelCase_ ( lowerCamelCase = None , lowerCamelCase = 128 ):
if function is None:
return functools.partial(lowerCamelCase , starting_batch_size=lowerCamelCase )
__magic_name__ : List[Any] =starting_batch_size
def decorator(*lowerCamelCase , **lowerCamelCase ):
nonlocal batch_size
gc.collect()
if is_xpu_available():
torch.xpu.empty_cache()
elif is_npu_available():
torch.npu.empty_cache()
else:
torch.cuda.empty_cache()
__magic_name__ : Optional[Any] =list(inspect.signature(lowerCamelCase ).parameters.keys() )
# Guard against user error
if len(lowerCamelCase ) < (len(lowerCamelCase ) + 1):
__magic_name__ : Optional[int] =""", """.join([F"{arg}={value}" for arg, value in zip(params[1:] , args[1:] )] )
raise TypeError(
F"Batch size was passed into `{function.__name__}` as the first argument when called."
F"Remove this as the decorator already does so: `{function.__name__}({arg_str})`" )
while True:
if batch_size == 0:
raise RuntimeError("""No executable batch size found, reached zero.""" )
try:
return function(lowerCamelCase , *lowerCamelCase , **lowerCamelCase )
except Exception as e:
if should_reduce_batch_size(lowerCamelCase ):
gc.collect()
if is_xpu_available():
torch.xpu.empty_cache()
elif is_npu_available():
torch.npu.empty_cache()
else:
torch.cuda.empty_cache()
batch_size //= 2
else:
raise
return decorator
| 21 |
import copy
from typing import TYPE_CHECKING, Any, Mapping, Optional, OrderedDict
from packaging import version
from ...configuration_utils import PretrainedConfig
from ...onnx import OnnxConfig
from ...utils import logging
from ..auto.configuration_auto import AutoConfig
if TYPE_CHECKING:
from ... import PreTrainedTokenizerBase, TensorType
snake_case__ : List[str] = logging.get_logger(__name__)
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
A_ = """vision-encoder-decoder"""
A_ = True
def __init__( self , **_UpperCAmelCase ) -> Dict:
super().__init__(**_UpperCAmelCase )
if "encoder" not in kwargs or "decoder" not in kwargs:
raise ValueError(
f"""A configuraton of type {self.model_type} cannot be instantiated because """
f"""not both `encoder` and `decoder` sub-configurations are passed, but only {kwargs}""" )
UpperCamelCase_ = kwargs.pop('encoder' )
UpperCamelCase_ = encoder_config.pop('model_type' )
UpperCamelCase_ = kwargs.pop('decoder' )
UpperCamelCase_ = decoder_config.pop('model_type' )
UpperCamelCase_ = AutoConfig.for_model(_UpperCAmelCase , **_UpperCAmelCase )
UpperCamelCase_ = AutoConfig.for_model(_UpperCAmelCase , **_UpperCAmelCase )
UpperCamelCase_ = True
@classmethod
def _UpperCAmelCase ( cls , _UpperCAmelCase , _UpperCAmelCase , **_UpperCAmelCase ) -> PretrainedConfig:
logger.info('Setting `config.is_decoder=True` and `config.add_cross_attention=True` for decoder_config' )
UpperCamelCase_ = True
UpperCamelCase_ = True
return cls(encoder=encoder_config.to_dict() , decoder=decoder_config.to_dict() , **_UpperCAmelCase )
def _UpperCAmelCase ( self ) -> int:
UpperCamelCase_ = copy.deepcopy(self.__dict__ )
UpperCamelCase_ = self.encoder.to_dict()
UpperCamelCase_ = self.decoder.to_dict()
UpperCamelCase_ = self.__class__.model_type
return output
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
A_ = version.parse("""1.11""" )
@property
def _UpperCAmelCase ( self ) -> Mapping[str, Mapping[int, str]]:
return OrderedDict(
[
('pixel_values', {0: 'batch', 1: 'num_channels', 2: 'height', 3: 'width'}),
] )
@property
def _UpperCAmelCase ( self ) -> float:
return 1e-4
@property
def _UpperCAmelCase ( self ) -> Mapping[str, Mapping[int, str]]:
return OrderedDict({'last_hidden_state': {0: 'batch', 1: 'encoder_sequence'}} )
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
@property
def _UpperCAmelCase ( self ) -> Mapping[str, Mapping[int, str]]:
UpperCamelCase_ = OrderedDict()
UpperCamelCase_ = {0: 'batch', 1: 'past_decoder_sequence + sequence'}
UpperCamelCase_ = {0: 'batch', 1: 'past_decoder_sequence + sequence'}
UpperCamelCase_ = {0: 'batch', 1: 'encoder_sequence'}
return common_inputs
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase = -1 , _UpperCAmelCase = -1 , _UpperCAmelCase = False , _UpperCAmelCase = None , ) -> Mapping[str, Any]:
import torch
UpperCamelCase_ = OrderedDict()
UpperCamelCase_ = super().generate_dummy_inputs(
_UpperCAmelCase , batch_size=_UpperCAmelCase , seq_length=_UpperCAmelCase , is_pair=_UpperCAmelCase , framework=_UpperCAmelCase )
UpperCamelCase_ , UpperCamelCase_ = dummy_input['input_ids'].shape
UpperCamelCase_ = (batch, encoder_sequence, self._config.encoder_hidden_size)
UpperCamelCase_ = dummy_input.pop('input_ids' )
UpperCamelCase_ = dummy_input.pop('attention_mask' )
UpperCamelCase_ = torch.zeros(_UpperCAmelCase )
return common_inputs
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
@property
def _UpperCAmelCase ( self ) -> None:
pass
def _UpperCAmelCase ( self , _UpperCAmelCase ) -> OnnxConfig:
return VisionEncoderDecoderEncoderOnnxConfig(_UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase = "default" ) -> OnnxConfig:
UpperCamelCase_ = encoder_config.hidden_size
return VisionEncoderDecoderDecoderOnnxConfig(_UpperCAmelCase , _UpperCAmelCase )
| 23 | 0 |
'''simple docstring'''
import logging
import os
import sys
from dataclasses import dataclass, field
from typing import Optional
import torch
from datasets import load_dataset
from torchvision.transforms import Compose, Lambda, Normalize, RandomHorizontalFlip, RandomResizedCrop, ToTensor
from torchvision.transforms.functional import InterpolationMode
import transformers
from transformers import (
HfArgumentParser,
Trainer,
TrainingArguments,
ViTImageProcessor,
ViTMAEConfig,
ViTMAEForPreTraining,
)
from transformers.trainer_utils import get_last_checkpoint
from transformers.utils import check_min_version, send_example_telemetry
from transformers.utils.versions import require_version
_snake_case : List[str] = logging.getLogger(__name__)
# Will error if the minimal version of Transformers is not installed. Remove at your own risks.
check_min_version('4.31.0')
require_version('datasets>=1.8.0', 'To fix: pip install -r examples/pytorch/image-pretraining/requirements.txt')
@dataclass
class A :
lowercase_ = field(
default='cifar10' ,metadata={'help': 'Name of a dataset from the datasets package'} )
lowercase_ = field(
default=_a ,metadata={'help': 'The configuration name of the dataset to use (via the datasets library).'} )
lowercase_ = field(
default=_a ,metadata={'help': 'The column name of the images in the files.'} )
lowercase_ = field(default=_a ,metadata={'help': 'A folder containing the training data.'} )
lowercase_ = field(default=_a ,metadata={'help': 'A folder containing the validation data.'} )
lowercase_ = field(
default=0.15 ,metadata={'help': 'Percent to split off of train for validation.'} )
lowercase_ = field(
default=_a ,metadata={
'help': (
'For debugging purposes or quicker training, truncate the number of training examples to this '
'value if set.'
)
} ,)
lowercase_ = field(
default=_a ,metadata={
'help': (
'For debugging purposes or quicker training, truncate the number of evaluation examples to this '
'value if set.'
)
} ,)
def __lowerCAmelCase ( self : int ) -> Any:
"""simple docstring"""
_a = {}
if self.train_dir is not None:
_a = self.train_dir
if self.validation_dir is not None:
_a = self.validation_dir
_a = data_files if data_files else None
@dataclass
class A :
lowercase_ = field(
default=_a ,metadata={
'help': (
'The model checkpoint for weights initialization.Don\'t set if you want to train a model from scratch.'
)
} ,)
lowercase_ = field(
default=_a ,metadata={'help': 'Pretrained config name or path if not the same as model_name_or_path'} )
lowercase_ = field(
default=_a ,metadata={
'help': (
'Override some existing default config settings when a model is trained from scratch. Example: '
'n_embd=10,resid_pdrop=0.2,scale_attn_weights=false,summary_type=cls_index'
)
} ,)
lowercase_ = field(
default=_a ,metadata={'help': 'Where do you want to store the pretrained models downloaded from s3'} )
lowercase_ = field(
default='main' ,metadata={'help': 'The specific model version to use (can be a branch name, tag name or commit id).'} ,)
lowercase_ = field(default=_a ,metadata={'help': 'Name or path of preprocessor config.'} )
lowercase_ = field(
default=_a ,metadata={
'help': (
'Will use the token generated when running `huggingface-cli login` (necessary to use this script '
'with private models).'
)
} ,)
lowercase_ = field(
default=0.75 ,metadata={'help': 'The ratio of the number of masked tokens in the input sequence.'} )
lowercase_ = field(
default=_a ,metadata={'help': 'Whether or not to train with normalized pixel values as target.'} )
@dataclass
class A ( _a ):
lowercase_ = field(
default=1e-3 ,metadata={'help': 'Base learning rate: absolute_lr = base_lr * total_batch_size / 256.'} )
def snake_case_ (UpperCamelCase : List[str] ):
'''simple docstring'''
_a = torch.stack([example['''pixel_values'''] for example in examples] )
return {"pixel_values": pixel_values}
def snake_case_ ():
'''simple docstring'''
_a = HfArgumentParser((ModelArguments, DataTrainingArguments, CustomTrainingArguments) )
if len(sys.argv ) == 2 and sys.argv[1].endswith('''.json''' ):
# If we pass only one argument to the script and it's the path to a json file,
# let's parse it to get our arguments.
_a , _a , _a = parser.parse_json_file(json_file=os.path.abspath(sys.argv[1] ) )
else:
_a , _a , _a = parser.parse_args_into_dataclasses()
# Sending telemetry. Tracking the example usage helps us better allocate resources to maintain them. The
# information sent is the one passed as arguments along with your Python/PyTorch versions.
send_example_telemetry('''run_mae''' , UpperCamelCase , UpperCamelCase )
# Setup logging
logging.basicConfig(
format='''%(asctime)s - %(levelname)s - %(name)s - %(message)s''' , datefmt='''%m/%d/%Y %H:%M:%S''' , handlers=[logging.StreamHandler(sys.stdout )] , )
if training_args.should_log:
# The default of training_args.log_level is passive, so we set log level at info here to have that default.
transformers.utils.logging.set_verbosity_info()
_a = training_args.get_process_log_level()
logger.setLevel(UpperCamelCase )
transformers.utils.logging.set_verbosity(UpperCamelCase )
transformers.utils.logging.enable_default_handler()
transformers.utils.logging.enable_explicit_format()
# Log on each process the small summary:
logger.warning(
f'Process rank: {training_args.local_rank}, device: {training_args.device}, n_gpu: {training_args.n_gpu}'
+ f'distributed training: {bool(training_args.local_rank != -1 )}, 16-bits training: {training_args.fpaa}' )
logger.info(f'Training/evaluation parameters {training_args}' )
# Detecting last checkpoint.
_a = None
if os.path.isdir(training_args.output_dir ) and training_args.do_train and not training_args.overwrite_output_dir:
_a = get_last_checkpoint(training_args.output_dir )
if last_checkpoint is None and len(os.listdir(training_args.output_dir ) ) > 0:
raise ValueError(
f'Output directory ({training_args.output_dir}) already exists and is not empty. '
'''Use --overwrite_output_dir to overcome.''' )
elif last_checkpoint is not None and training_args.resume_from_checkpoint is None:
logger.info(
f'Checkpoint detected, resuming training at {last_checkpoint}. To avoid this behavior, change '
'''the `--output_dir` or add `--overwrite_output_dir` to train from scratch.''' )
# Initialize our dataset.
_a = load_dataset(
data_args.dataset_name , data_args.dataset_config_name , data_files=data_args.data_files , cache_dir=model_args.cache_dir , use_auth_token=True if model_args.use_auth_token else None , )
# If we don't have a validation split, split off a percentage of train as validation.
_a = None if '''validation''' in ds.keys() else data_args.train_val_split
if isinstance(data_args.train_val_split , UpperCamelCase ) and data_args.train_val_split > 0.0:
_a = ds['''train'''].train_test_split(data_args.train_val_split )
_a = split['''train''']
_a = split['''test''']
# Load pretrained model and image processor
#
# Distributed training:
# The .from_pretrained methods guarantee that only one local process can concurrently
# download model & vocab.
_a = {
'''cache_dir''': model_args.cache_dir,
'''revision''': model_args.model_revision,
'''use_auth_token''': True if model_args.use_auth_token else None,
}
if model_args.config_name:
_a = ViTMAEConfig.from_pretrained(model_args.config_name , **UpperCamelCase )
elif model_args.model_name_or_path:
_a = ViTMAEConfig.from_pretrained(model_args.model_name_or_path , **UpperCamelCase )
else:
_a = ViTMAEConfig()
logger.warning('''You are instantiating a new config instance from scratch.''' )
if model_args.config_overrides is not None:
logger.info(f'Overriding config: {model_args.config_overrides}' )
config.update_from_string(model_args.config_overrides )
logger.info(f'New config: {config}' )
# adapt config
config.update(
{
'''mask_ratio''': model_args.mask_ratio,
'''norm_pix_loss''': model_args.norm_pix_loss,
} )
# create image processor
if model_args.image_processor_name:
_a = ViTImageProcessor.from_pretrained(model_args.image_processor_name , **UpperCamelCase )
elif model_args.model_name_or_path:
_a = ViTImageProcessor.from_pretrained(model_args.model_name_or_path , **UpperCamelCase )
else:
_a = ViTImageProcessor()
# create model
if model_args.model_name_or_path:
_a = ViTMAEForPreTraining.from_pretrained(
model_args.model_name_or_path , from_tf=bool('''.ckpt''' in model_args.model_name_or_path ) , config=UpperCamelCase , cache_dir=model_args.cache_dir , revision=model_args.model_revision , use_auth_token=True if model_args.use_auth_token else None , )
else:
logger.info('''Training new model from scratch''' )
_a = ViTMAEForPreTraining(UpperCamelCase )
if training_args.do_train:
_a = ds['''train'''].column_names
else:
_a = ds['''validation'''].column_names
if data_args.image_column_name is not None:
_a = data_args.image_column_name
elif "image" in column_names:
_a = '''image'''
elif "img" in column_names:
_a = '''img'''
else:
_a = column_names[0]
# transformations as done in original MAE paper
# source: https://github.com/facebookresearch/mae/blob/main/main_pretrain.py
if "shortest_edge" in image_processor.size:
_a = image_processor.size['''shortest_edge''']
else:
_a = (image_processor.size['''height'''], image_processor.size['''width'''])
_a = Compose(
[
Lambda(lambda UpperCamelCase : img.convert('''RGB''' ) if img.mode != "RGB" else img ),
RandomResizedCrop(UpperCamelCase , scale=(0.2, 1.0) , interpolation=InterpolationMode.BICUBIC ),
RandomHorizontalFlip(),
ToTensor(),
Normalize(mean=image_processor.image_mean , std=image_processor.image_std ),
] )
def preprocess_images(UpperCamelCase : Union[str, Any] ):
_a = [transforms(UpperCamelCase ) for image in examples[image_column_name]]
return examples
if training_args.do_train:
if "train" not in ds:
raise ValueError('''--do_train requires a train dataset''' )
if data_args.max_train_samples is not None:
_a = ds['''train'''].shuffle(seed=training_args.seed ).select(range(data_args.max_train_samples ) )
# Set the training transforms
ds["train"].set_transform(UpperCamelCase )
if training_args.do_eval:
if "validation" not in ds:
raise ValueError('''--do_eval requires a validation dataset''' )
if data_args.max_eval_samples is not None:
_a = (
ds['''validation'''].shuffle(seed=training_args.seed ).select(range(data_args.max_eval_samples ) )
)
# Set the validation transforms
ds["validation"].set_transform(UpperCamelCase )
# Compute absolute learning rate
_a = (
training_args.train_batch_size * training_args.gradient_accumulation_steps * training_args.world_size
)
if training_args.base_learning_rate is not None:
_a = training_args.base_learning_rate * total_train_batch_size / 256
# Initialize our trainer
_a = Trainer(
model=UpperCamelCase , args=UpperCamelCase , train_dataset=ds['''train'''] if training_args.do_train else None , eval_dataset=ds['''validation'''] if training_args.do_eval else None , tokenizer=UpperCamelCase , data_collator=UpperCamelCase , )
# Training
if training_args.do_train:
_a = None
if training_args.resume_from_checkpoint is not None:
_a = training_args.resume_from_checkpoint
elif last_checkpoint is not None:
_a = last_checkpoint
_a = trainer.train(resume_from_checkpoint=UpperCamelCase )
trainer.save_model()
trainer.log_metrics('''train''' , train_result.metrics )
trainer.save_metrics('''train''' , train_result.metrics )
trainer.save_state()
# Evaluation
if training_args.do_eval:
_a = trainer.evaluate()
trainer.log_metrics('''eval''' , UpperCamelCase )
trainer.save_metrics('''eval''' , UpperCamelCase )
# Write model card and (optionally) push to hub
_a = {
'''tasks''': '''masked-auto-encoding''',
'''dataset''': data_args.dataset_name,
'''tags''': ['''masked-auto-encoding'''],
}
if training_args.push_to_hub:
trainer.push_to_hub(**UpperCamelCase )
else:
trainer.create_model_card(**UpperCamelCase )
def snake_case_ (UpperCamelCase : Optional[int] ):
'''simple docstring'''
main()
if __name__ == "__main__":
main()
| 22 |
import argparse
import torch
from transformers import MobileBertConfig, MobileBertForPreTraining, load_tf_weights_in_mobilebert
from transformers.utils import logging
logging.set_verbosity_info()
def _snake_case (__lowercase , __lowercase , __lowercase):
# Initialise PyTorch model
UpperCamelCase_ = MobileBertConfig.from_json_file(__lowercase)
print(f"""Building PyTorch model from configuration: {config}""")
UpperCamelCase_ = MobileBertForPreTraining(__lowercase)
# Load weights from tf checkpoint
UpperCamelCase_ = load_tf_weights_in_mobilebert(__lowercase , __lowercase , __lowercase)
# Save pytorch-model
print(f"""Save PyTorch model to {pytorch_dump_path}""")
torch.save(model.state_dict() , __lowercase)
if __name__ == "__main__":
snake_case__ : int = argparse.ArgumentParser()
# Required parameters
parser.add_argument(
"""--tf_checkpoint_path""", default=None, type=str, required=True, help="""Path to the TensorFlow checkpoint path."""
)
parser.add_argument(
"""--mobilebert_config_file""",
default=None,
type=str,
required=True,
help=(
"""The config json file corresponding to the pre-trained MobileBERT model. \n"""
"""This specifies the model architecture."""
),
)
parser.add_argument(
"""--pytorch_dump_path""", default=None, type=str, required=True, help="""Path to the output PyTorch model."""
)
snake_case__ : Optional[Any] = parser.parse_args()
convert_tf_checkpoint_to_pytorch(args.tf_checkpoint_path, args.mobilebert_config_file, args.pytorch_dump_path)
| 23 | 0 |
'''simple docstring'''
import json
import os
import shutil
import tempfile
import unittest
from multiprocessing import get_context
from pathlib import Path
import datasets
import numpy as np
from datasets import load_dataset
from parameterized import parameterized
from transformers import AutoProcessor
from transformers.models.wavaveca import WavaVecaCTCTokenizer, WavaVecaFeatureExtractor
from transformers.models.wavaveca.tokenization_wavaveca import VOCAB_FILES_NAMES
from transformers.testing_utils import require_pyctcdecode, require_torch, require_torchaudio, slow
from transformers.utils import FEATURE_EXTRACTOR_NAME, is_pyctcdecode_available, is_torch_available
from ..wavaveca.test_feature_extraction_wavaveca import floats_list
if is_pyctcdecode_available():
from huggingface_hub import snapshot_download
from pyctcdecode import BeamSearchDecoderCTC
from transformers.models.wavaveca_with_lm import WavaVecaProcessorWithLM
from transformers.models.wavaveca_with_lm.processing_wavaveca_with_lm import WavaVecaDecoderWithLMOutput
if is_torch_available():
from transformers import WavaVecaForCTC
@require_pyctcdecode
class lowerCAmelCase ( unittest.TestCase):
def lowerCAmelCase ( self ) -> Any:
'''simple docstring'''
__snake_case = '''| <pad> <unk> <s> </s> a b c d e f g h i j k'''.split()
__snake_case = dict(zip(__SCREAMING_SNAKE_CASE , range(len(__SCREAMING_SNAKE_CASE ) ) ) )
__snake_case = {
'''unk_token''': '''<unk>''',
'''bos_token''': '''<s>''',
'''eos_token''': '''</s>''',
}
__snake_case = {
'''feature_size''': 1,
'''padding_value''': 0.0,
'''sampling_rate''': 1_6000,
'''return_attention_mask''': False,
'''do_normalize''': True,
}
__snake_case = tempfile.mkdtemp()
__snake_case = os.path.join(self.tmpdirname , VOCAB_FILES_NAMES['''vocab_file'''] )
__snake_case = os.path.join(self.tmpdirname , __SCREAMING_SNAKE_CASE )
with open(self.vocab_file , '''w''' , encoding='''utf-8''' ) as fp:
fp.write(json.dumps(__SCREAMING_SNAKE_CASE ) + '''\n''' )
with open(self.feature_extraction_file , '''w''' , encoding='''utf-8''' ) as fp:
fp.write(json.dumps(__SCREAMING_SNAKE_CASE ) + '''\n''' )
# load decoder from hub
__snake_case = '''hf-internal-testing/ngram-beam-search-decoder'''
def lowerCAmelCase ( self , **__SCREAMING_SNAKE_CASE ) -> Dict:
'''simple docstring'''
__snake_case = self.add_kwargs_tokens_map.copy()
kwargs.update(__SCREAMING_SNAKE_CASE )
return WavaVecaCTCTokenizer.from_pretrained(self.tmpdirname , **__SCREAMING_SNAKE_CASE )
def lowerCAmelCase ( self , **__SCREAMING_SNAKE_CASE ) -> Tuple:
'''simple docstring'''
return WavaVecaFeatureExtractor.from_pretrained(self.tmpdirname , **__SCREAMING_SNAKE_CASE )
def lowerCAmelCase ( self , **__SCREAMING_SNAKE_CASE ) -> Optional[int]:
'''simple docstring'''
return BeamSearchDecoderCTC.load_from_hf_hub(self.decoder_name , **__SCREAMING_SNAKE_CASE )
def lowerCAmelCase ( self ) -> int:
'''simple docstring'''
shutil.rmtree(self.tmpdirname )
def lowerCAmelCase ( self ) -> Dict:
'''simple docstring'''
__snake_case = self.get_tokenizer()
__snake_case = self.get_feature_extractor()
__snake_case = self.get_decoder()
__snake_case = WavaVecaProcessorWithLM(tokenizer=__SCREAMING_SNAKE_CASE , feature_extractor=__SCREAMING_SNAKE_CASE , decoder=__SCREAMING_SNAKE_CASE )
processor.save_pretrained(self.tmpdirname )
__snake_case = WavaVecaProcessorWithLM.from_pretrained(self.tmpdirname )
# tokenizer
self.assertEqual(processor.tokenizer.get_vocab() , tokenizer.get_vocab() )
self.assertIsInstance(processor.tokenizer , __SCREAMING_SNAKE_CASE )
# feature extractor
self.assertEqual(processor.feature_extractor.to_json_string() , feature_extractor.to_json_string() )
self.assertIsInstance(processor.feature_extractor , __SCREAMING_SNAKE_CASE )
# decoder
self.assertEqual(processor.decoder._alphabet.labels , decoder._alphabet.labels )
self.assertEqual(
processor.decoder.model_container[decoder._model_key]._unigram_set , decoder.model_container[decoder._model_key]._unigram_set , )
self.assertIsInstance(processor.decoder , __SCREAMING_SNAKE_CASE )
def lowerCAmelCase ( self ) -> Dict:
'''simple docstring'''
__snake_case = WavaVecaProcessorWithLM(
tokenizer=self.get_tokenizer() , feature_extractor=self.get_feature_extractor() , decoder=self.get_decoder() )
processor.save_pretrained(self.tmpdirname )
# make sure that error is thrown when decoder alphabet doesn't match
__snake_case = WavaVecaProcessorWithLM.from_pretrained(
self.tmpdirname , alpha=5.0 , beta=3.0 , score_boundary=-7.0 , unk_score_offset=3 )
# decoder
self.assertEqual(processor.language_model.alpha , 5.0 )
self.assertEqual(processor.language_model.beta , 3.0 )
self.assertEqual(processor.language_model.score_boundary , -7.0 )
self.assertEqual(processor.language_model.unk_score_offset , 3 )
def lowerCAmelCase ( self ) -> List[Any]:
'''simple docstring'''
__snake_case = self.get_tokenizer()
# add token to trigger raise
tokenizer.add_tokens(['''xx'''] )
with self.assertRaisesRegex(__SCREAMING_SNAKE_CASE , '''include''' ):
WavaVecaProcessorWithLM(
tokenizer=__SCREAMING_SNAKE_CASE , feature_extractor=self.get_feature_extractor() , decoder=self.get_decoder() )
def lowerCAmelCase ( self ) -> Union[str, Any]:
'''simple docstring'''
__snake_case = self.get_feature_extractor()
__snake_case = self.get_tokenizer()
__snake_case = self.get_decoder()
__snake_case = WavaVecaProcessorWithLM(tokenizer=__SCREAMING_SNAKE_CASE , feature_extractor=__SCREAMING_SNAKE_CASE , decoder=__SCREAMING_SNAKE_CASE )
__snake_case = floats_list((3, 1000) )
__snake_case = feature_extractor(__SCREAMING_SNAKE_CASE , return_tensors='''np''' )
__snake_case = processor(__SCREAMING_SNAKE_CASE , return_tensors='''np''' )
for key in input_feat_extract.keys():
self.assertAlmostEqual(input_feat_extract[key].sum() , input_processor[key].sum() , delta=1E-2 )
def lowerCAmelCase ( self ) -> Tuple:
'''simple docstring'''
__snake_case = self.get_feature_extractor()
__snake_case = self.get_tokenizer()
__snake_case = self.get_decoder()
__snake_case = WavaVecaProcessorWithLM(tokenizer=__SCREAMING_SNAKE_CASE , feature_extractor=__SCREAMING_SNAKE_CASE , decoder=__SCREAMING_SNAKE_CASE )
__snake_case = '''This is a test string'''
__snake_case = processor(text=__SCREAMING_SNAKE_CASE )
__snake_case = tokenizer(__SCREAMING_SNAKE_CASE )
for key in encoded_tok.keys():
self.assertListEqual(encoded_tok[key] , encoded_processor[key] )
def lowerCAmelCase ( self , __SCREAMING_SNAKE_CASE=(2, 10, 16) , __SCREAMING_SNAKE_CASE=77 ) -> Tuple:
'''simple docstring'''
np.random.seed(__SCREAMING_SNAKE_CASE )
return np.random.rand(*__SCREAMING_SNAKE_CASE )
def lowerCAmelCase ( self ) -> Optional[Any]:
'''simple docstring'''
__snake_case = self.get_feature_extractor()
__snake_case = self.get_tokenizer()
__snake_case = self.get_decoder()
__snake_case = WavaVecaProcessorWithLM(tokenizer=__SCREAMING_SNAKE_CASE , feature_extractor=__SCREAMING_SNAKE_CASE , decoder=__SCREAMING_SNAKE_CASE )
__snake_case = self._get_dummy_logits(shape=(10, 16) , seed=13 )
__snake_case = processor.decode(__SCREAMING_SNAKE_CASE )
__snake_case = decoder.decode_beams(__SCREAMING_SNAKE_CASE )[0]
self.assertEqual(decoded_decoder[0] , decoded_processor.text )
self.assertEqual('''</s> <s> </s>''' , decoded_processor.text )
self.assertEqual(decoded_decoder[-2] , decoded_processor.logit_score )
self.assertEqual(decoded_decoder[-1] , decoded_processor.lm_score )
@parameterized.expand([[None], ['''fork'''], ['''spawn''']] )
def lowerCAmelCase ( self , __SCREAMING_SNAKE_CASE ) -> Dict:
'''simple docstring'''
__snake_case = self.get_feature_extractor()
__snake_case = self.get_tokenizer()
__snake_case = self.get_decoder()
__snake_case = WavaVecaProcessorWithLM(tokenizer=__SCREAMING_SNAKE_CASE , feature_extractor=__SCREAMING_SNAKE_CASE , decoder=__SCREAMING_SNAKE_CASE )
__snake_case = self._get_dummy_logits()
# note: pool should be instantiated *after* Wav2Vec2ProcessorWithLM.
# otherwise, the LM won't be available to the pool's sub-processes.
# manual logic used to allow parameterized test for both pool=None and pool=Pool(...)
if pool_context is None:
__snake_case = processor.batch_decode(__SCREAMING_SNAKE_CASE )
else:
with get_context(__SCREAMING_SNAKE_CASE ).Pool() as pool:
__snake_case = processor.batch_decode(__SCREAMING_SNAKE_CASE , __SCREAMING_SNAKE_CASE )
__snake_case = list(__SCREAMING_SNAKE_CASE )
with get_context('''fork''' ).Pool() as p:
__snake_case = decoder.decode_beams_batch(__SCREAMING_SNAKE_CASE , __SCREAMING_SNAKE_CASE )
__snake_case , __snake_case , __snake_case = [], [], []
for beams in decoded_beams:
texts_decoder.append(beams[0][0] )
logit_scores_decoder.append(beams[0][-2] )
lm_scores_decoder.append(beams[0][-1] )
self.assertListEqual(__SCREAMING_SNAKE_CASE , decoded_processor.text )
self.assertListEqual(['''<s> <s> </s>''', '''<s> <s> <s>'''] , decoded_processor.text )
self.assertListEqual(__SCREAMING_SNAKE_CASE , decoded_processor.logit_score )
self.assertListEqual(__SCREAMING_SNAKE_CASE , decoded_processor.lm_score )
def lowerCAmelCase ( self ) -> Any:
'''simple docstring'''
__snake_case = self.get_feature_extractor()
__snake_case = self.get_tokenizer()
__snake_case = self.get_decoder()
__snake_case = WavaVecaProcessorWithLM(tokenizer=__SCREAMING_SNAKE_CASE , feature_extractor=__SCREAMING_SNAKE_CASE , decoder=__SCREAMING_SNAKE_CASE )
__snake_case = self._get_dummy_logits()
__snake_case = 15
__snake_case = -20.0
__snake_case = -4.0
__snake_case = processor.batch_decode(
__SCREAMING_SNAKE_CASE , beam_width=__SCREAMING_SNAKE_CASE , beam_prune_logp=__SCREAMING_SNAKE_CASE , token_min_logp=__SCREAMING_SNAKE_CASE , )
__snake_case = decoded_processor_out.text
__snake_case = list(__SCREAMING_SNAKE_CASE )
with get_context('''fork''' ).Pool() as pool:
__snake_case = decoder.decode_beams_batch(
__SCREAMING_SNAKE_CASE , __SCREAMING_SNAKE_CASE , beam_width=__SCREAMING_SNAKE_CASE , beam_prune_logp=__SCREAMING_SNAKE_CASE , token_min_logp=__SCREAMING_SNAKE_CASE , )
__snake_case = [d[0][0] for d in decoded_decoder_out]
__snake_case = [d[0][2] for d in decoded_decoder_out]
__snake_case = [d[0][3] for d in decoded_decoder_out]
self.assertListEqual(__SCREAMING_SNAKE_CASE , __SCREAMING_SNAKE_CASE )
self.assertListEqual(['''</s> <s> <s>''', '''<s> <s> <s>'''] , __SCREAMING_SNAKE_CASE )
self.assertTrue(np.array_equal(__SCREAMING_SNAKE_CASE , decoded_processor_out.logit_score ) )
self.assertTrue(np.allclose([-20.054, -18.447] , __SCREAMING_SNAKE_CASE , atol=1E-3 ) )
self.assertTrue(np.array_equal(__SCREAMING_SNAKE_CASE , decoded_processor_out.lm_score ) )
self.assertTrue(np.allclose([-15.554, -13.9_474] , __SCREAMING_SNAKE_CASE , atol=1E-3 ) )
def lowerCAmelCase ( self ) -> Optional[Any]:
'''simple docstring'''
__snake_case = self.get_feature_extractor()
__snake_case = self.get_tokenizer()
__snake_case = self.get_decoder()
__snake_case = WavaVecaProcessorWithLM(tokenizer=__SCREAMING_SNAKE_CASE , feature_extractor=__SCREAMING_SNAKE_CASE , decoder=__SCREAMING_SNAKE_CASE )
__snake_case = self._get_dummy_logits()
__snake_case = 2.0
__snake_case = 5.0
__snake_case = -20.0
__snake_case = True
__snake_case = processor.batch_decode(
__SCREAMING_SNAKE_CASE , alpha=__SCREAMING_SNAKE_CASE , beta=__SCREAMING_SNAKE_CASE , unk_score_offset=__SCREAMING_SNAKE_CASE , lm_score_boundary=__SCREAMING_SNAKE_CASE , )
__snake_case = decoded_processor_out.text
__snake_case = list(__SCREAMING_SNAKE_CASE )
decoder.reset_params(
alpha=__SCREAMING_SNAKE_CASE , beta=__SCREAMING_SNAKE_CASE , unk_score_offset=__SCREAMING_SNAKE_CASE , lm_score_boundary=__SCREAMING_SNAKE_CASE , )
with get_context('''fork''' ).Pool() as pool:
__snake_case = decoder.decode_beams_batch(
__SCREAMING_SNAKE_CASE , __SCREAMING_SNAKE_CASE , )
__snake_case = [d[0][0] for d in decoded_decoder_out]
self.assertListEqual(__SCREAMING_SNAKE_CASE , __SCREAMING_SNAKE_CASE )
self.assertListEqual(['''<s> </s> <s> </s> </s>''', '''</s> </s> <s> </s> </s>'''] , __SCREAMING_SNAKE_CASE )
__snake_case = processor.decoder.model_container[processor.decoder._model_key]
self.assertEqual(lm_model.alpha , 2.0 )
self.assertEqual(lm_model.beta , 5.0 )
self.assertEqual(lm_model.unk_score_offset , -20.0 )
self.assertEqual(lm_model.score_boundary , __SCREAMING_SNAKE_CASE )
def lowerCAmelCase ( self ) -> Union[str, Any]:
'''simple docstring'''
__snake_case = WavaVecaProcessorWithLM.from_pretrained('''hf-internal-testing/processor_with_lm''' )
__snake_case = processor.decoder.model_container[processor.decoder._model_key]
__snake_case = Path(language_model._kenlm_model.path.decode('''utf-8''' ) ).parent.parent.absolute()
__snake_case = os.listdir(__SCREAMING_SNAKE_CASE )
__snake_case = ['''alphabet.json''', '''language_model''']
downloaded_decoder_files.sort()
expected_decoder_files.sort()
# test that only decoder relevant files from
# https://huggingface.co/hf-internal-testing/processor_with_lm/tree/main
# are downloaded and none of the rest (e.g. README.md, ...)
self.assertListEqual(__SCREAMING_SNAKE_CASE , __SCREAMING_SNAKE_CASE )
def lowerCAmelCase ( self ) -> Any:
'''simple docstring'''
__snake_case = snapshot_download('''hf-internal-testing/processor_with_lm''' )
__snake_case = WavaVecaProcessorWithLM.from_pretrained(__SCREAMING_SNAKE_CASE )
__snake_case = processor.decoder.model_container[processor.decoder._model_key]
__snake_case = Path(language_model._kenlm_model.path.decode('''utf-8''' ) ).parent.parent.absolute()
__snake_case = os.listdir(__SCREAMING_SNAKE_CASE )
__snake_case = os.listdir(__SCREAMING_SNAKE_CASE )
local_decoder_files.sort()
expected_decoder_files.sort()
# test that both decoder form hub and local files in cache are the same
self.assertListEqual(__SCREAMING_SNAKE_CASE , __SCREAMING_SNAKE_CASE )
def lowerCAmelCase ( self ) -> Union[str, Any]:
'''simple docstring'''
__snake_case = WavaVecaProcessorWithLM.from_pretrained('''hf-internal-testing/processor_with_lm''' )
__snake_case = AutoProcessor.from_pretrained('''hf-internal-testing/processor_with_lm''' )
__snake_case = floats_list((3, 1000) )
__snake_case = processor_wavaveca(__SCREAMING_SNAKE_CASE , return_tensors='''np''' )
__snake_case = processor_auto(__SCREAMING_SNAKE_CASE , return_tensors='''np''' )
for key in input_wavaveca.keys():
self.assertAlmostEqual(input_wavaveca[key].sum() , input_auto[key].sum() , delta=1E-2 )
__snake_case = self._get_dummy_logits()
__snake_case = processor_wavaveca.batch_decode(__SCREAMING_SNAKE_CASE )
__snake_case = processor_auto.batch_decode(__SCREAMING_SNAKE_CASE )
self.assertListEqual(decoded_wavaveca.text , decoded_auto.text )
def lowerCAmelCase ( self ) -> List[str]:
'''simple docstring'''
__snake_case = self.get_feature_extractor()
__snake_case = self.get_tokenizer()
__snake_case = self.get_decoder()
__snake_case = WavaVecaProcessorWithLM(tokenizer=__SCREAMING_SNAKE_CASE , feature_extractor=__SCREAMING_SNAKE_CASE , decoder=__SCREAMING_SNAKE_CASE )
self.assertListEqual(
processor.model_input_names , feature_extractor.model_input_names , msg='''`processor` and `feature_extractor` model input names do not match''' , )
@staticmethod
def lowerCAmelCase ( __SCREAMING_SNAKE_CASE , __SCREAMING_SNAKE_CASE ) -> Dict:
'''simple docstring'''
__snake_case = [d[key] for d in offsets]
return retrieved_list
def lowerCAmelCase ( self ) -> Dict:
'''simple docstring'''
__snake_case = WavaVecaProcessorWithLM.from_pretrained('''hf-internal-testing/processor_with_lm''' )
__snake_case = self._get_dummy_logits()[0]
__snake_case = processor.decode(__SCREAMING_SNAKE_CASE , output_word_offsets=__SCREAMING_SNAKE_CASE )
# check Wav2Vec2CTCTokenizerOutput keys for word
self.assertEqual(len(outputs.keys() ) , 4 )
self.assertTrue('''text''' in outputs )
self.assertTrue('''word_offsets''' in outputs )
self.assertTrue(isinstance(__SCREAMING_SNAKE_CASE , __SCREAMING_SNAKE_CASE ) )
self.assertEqual(''' '''.join(self.get_from_offsets(outputs['''word_offsets'''] , '''word''' ) ) , outputs.text )
self.assertListEqual(self.get_from_offsets(outputs['''word_offsets'''] , '''word''' ) , ['''<s>''', '''<s>''', '''</s>'''] )
self.assertListEqual(self.get_from_offsets(outputs['''word_offsets'''] , '''start_offset''' ) , [0, 2, 4] )
self.assertListEqual(self.get_from_offsets(outputs['''word_offsets'''] , '''end_offset''' ) , [1, 3, 5] )
def lowerCAmelCase ( self ) -> str:
'''simple docstring'''
__snake_case = WavaVecaProcessorWithLM.from_pretrained('''hf-internal-testing/processor_with_lm''' )
__snake_case = self._get_dummy_logits()
__snake_case = processor.batch_decode(__SCREAMING_SNAKE_CASE , output_word_offsets=__SCREAMING_SNAKE_CASE )
# check Wav2Vec2CTCTokenizerOutput keys for word
self.assertEqual(len(outputs.keys() ) , 4 )
self.assertTrue('''text''' in outputs )
self.assertTrue('''word_offsets''' in outputs )
self.assertTrue(isinstance(__SCREAMING_SNAKE_CASE , __SCREAMING_SNAKE_CASE ) )
self.assertListEqual(
[''' '''.join(self.get_from_offsets(__SCREAMING_SNAKE_CASE , '''word''' ) ) for o in outputs['''word_offsets''']] , outputs.text )
self.assertListEqual(self.get_from_offsets(outputs['''word_offsets'''][0] , '''word''' ) , ['''<s>''', '''<s>''', '''</s>'''] )
self.assertListEqual(self.get_from_offsets(outputs['''word_offsets'''][0] , '''start_offset''' ) , [0, 2, 4] )
self.assertListEqual(self.get_from_offsets(outputs['''word_offsets'''][0] , '''end_offset''' ) , [1, 3, 5] )
@slow
@require_torch
@require_torchaudio
def lowerCAmelCase ( self ) -> Union[str, Any]:
'''simple docstring'''
import torch
__snake_case = load_dataset('''common_voice''' , '''en''' , split='''train''' , streaming=__SCREAMING_SNAKE_CASE )
__snake_case = ds.cast_column('''audio''' , datasets.Audio(sampling_rate=1_6000 ) )
__snake_case = iter(__SCREAMING_SNAKE_CASE )
__snake_case = next(__SCREAMING_SNAKE_CASE )
__snake_case = AutoProcessor.from_pretrained('''patrickvonplaten/wav2vec2-base-100h-with-lm''' )
__snake_case = WavaVecaForCTC.from_pretrained('''patrickvonplaten/wav2vec2-base-100h-with-lm''' )
# compare to filename `common_voice_en_100038.mp3` of dataset viewer on https://huggingface.co/datasets/common_voice/viewer/en/train
__snake_case = processor(sample['''audio''']['''array'''] , return_tensors='''pt''' ).input_values
with torch.no_grad():
__snake_case = model(__SCREAMING_SNAKE_CASE ).logits.cpu().numpy()
__snake_case = processor.decode(logits[0] , output_word_offsets=__SCREAMING_SNAKE_CASE )
__snake_case = model.config.inputs_to_logits_ratio / processor.feature_extractor.sampling_rate
__snake_case = [
{
'''start_time''': d['''start_offset'''] * time_offset,
'''end_time''': d['''end_offset'''] * time_offset,
'''word''': d['''word'''],
}
for d in output['''word_offsets''']
]
__snake_case = '''WHY DOES MILISANDRA LOOK LIKE SHE WANTS TO CONSUME JOHN SNOW ON THE RIVER AT THE WALL'''
# output words
self.assertEqual(''' '''.join(self.get_from_offsets(__SCREAMING_SNAKE_CASE , '''word''' ) ) , __SCREAMING_SNAKE_CASE )
self.assertEqual(''' '''.join(self.get_from_offsets(__SCREAMING_SNAKE_CASE , '''word''' ) ) , output.text )
# output times
__snake_case = torch.tensor(self.get_from_offsets(__SCREAMING_SNAKE_CASE , '''start_time''' ) )
__snake_case = torch.tensor(self.get_from_offsets(__SCREAMING_SNAKE_CASE , '''end_time''' ) )
# fmt: off
__snake_case = torch.tensor([1.4_199, 1.6_599, 2.2_599, 3.0, 3.24, 3.5_999, 3.7_999, 4.0_999, 4.26, 4.94, 5.28, 5.6_599, 5.78, 5.94, 6.32, 6.5_399, 6.6_599] )
__snake_case = torch.tensor([1.5_399, 1.8_999, 2.9, 3.16, 3.5_399, 3.72, 4.0_199, 4.1_799, 4.76, 5.1_599, 5.5_599, 5.6_999, 5.86, 6.1_999, 6.38, 6.6_199, 6.94] )
# fmt: on
self.assertTrue(torch.allclose(__SCREAMING_SNAKE_CASE , __SCREAMING_SNAKE_CASE , atol=0.01 ) )
self.assertTrue(torch.allclose(__SCREAMING_SNAKE_CASE , __SCREAMING_SNAKE_CASE , atol=0.01 ) )
| 24 |
import gc
import unittest
from transformers import MODEL_FOR_MASKED_LM_MAPPING, TF_MODEL_FOR_MASKED_LM_MAPPING, FillMaskPipeline, pipeline
from transformers.pipelines import PipelineException
from transformers.testing_utils import (
is_pipeline_test,
is_torch_available,
nested_simplify,
require_tf,
require_torch,
require_torch_gpu,
slow,
)
from .test_pipelines_common import ANY
@is_pipeline_test
class _a ( unittest.TestCase ):
"""simple docstring"""
A_ = MODEL_FOR_MASKED_LM_MAPPING
A_ = TF_MODEL_FOR_MASKED_LM_MAPPING
def _UpperCAmelCase ( self ) -> List[str]:
super().tearDown()
# clean-up as much as possible GPU memory occupied by PyTorch
gc.collect()
if is_torch_available():
import torch
torch.cuda.empty_cache()
@require_tf
def _UpperCAmelCase ( self ) -> str:
UpperCamelCase_ = pipeline(task='fill-mask' , model='sshleifer/tiny-distilroberta-base' , top_k=2 , framework='tf' )
UpperCamelCase_ = unmasker('My name is <mask>' )
self.assertEqual(
nested_simplify(_UpperCAmelCase , decimals=6 ) , [
{'sequence': 'My name is grouped', 'score': 2.1e-05, 'token': 38015, 'token_str': ' grouped'},
{'sequence': 'My name is accuser', 'score': 2.1e-05, 'token': 25506, 'token_str': ' accuser'},
] , )
UpperCamelCase_ = unmasker('The largest city in France is <mask>' )
self.assertEqual(
nested_simplify(_UpperCAmelCase , decimals=6 ) , [
{
'sequence': 'The largest city in France is grouped',
'score': 2.1e-05,
'token': 38015,
'token_str': ' grouped',
},
{
'sequence': 'The largest city in France is accuser',
'score': 2.1e-05,
'token': 25506,
'token_str': ' accuser',
},
] , )
UpperCamelCase_ = unmasker('My name is <mask>' , targets=[' Patrick', ' Clara', ' Teven'] , top_k=3 )
self.assertEqual(
nested_simplify(_UpperCAmelCase , decimals=6 ) , [
{'sequence': 'My name is Clara', 'score': 2e-05, 'token': 13606, 'token_str': ' Clara'},
{'sequence': 'My name is Patrick', 'score': 2e-05, 'token': 3499, 'token_str': ' Patrick'},
{'sequence': 'My name is Te', 'score': 1.9e-05, 'token': 2941, 'token_str': ' Te'},
] , )
@require_torch
def _UpperCAmelCase ( self ) -> Union[str, Any]:
UpperCamelCase_ = pipeline(task='fill-mask' , model='sshleifer/tiny-distilroberta-base' , top_k=2 , framework='pt' )
UpperCamelCase_ = unmasker('My name is <mask>' )
self.assertEqual(
nested_simplify(_UpperCAmelCase , decimals=6 ) , [
{'sequence': 'My name is Maul', 'score': 2.2e-05, 'token': 35676, 'token_str': ' Maul'},
{'sequence': 'My name isELS', 'score': 2.2e-05, 'token': 16416, 'token_str': 'ELS'},
] , )
UpperCamelCase_ = unmasker('The largest city in France is <mask>' )
self.assertEqual(
nested_simplify(_UpperCAmelCase , decimals=6 ) , [
{
'sequence': 'The largest city in France is Maul',
'score': 2.2e-05,
'token': 35676,
'token_str': ' Maul',
},
{'sequence': 'The largest city in France isELS', 'score': 2.2e-05, 'token': 16416, 'token_str': 'ELS'},
] , )
UpperCamelCase_ = unmasker('My name is <mask>' , targets=[' Patrick', ' Clara', ' Teven'] , top_k=3 )
self.assertEqual(
nested_simplify(_UpperCAmelCase , decimals=6 ) , [
{'sequence': 'My name is Patrick', 'score': 2.1e-05, 'token': 3499, 'token_str': ' Patrick'},
{'sequence': 'My name is Te', 'score': 2e-05, 'token': 2941, 'token_str': ' Te'},
{'sequence': 'My name is Clara', 'score': 2e-05, 'token': 13606, 'token_str': ' Clara'},
] , )
UpperCamelCase_ = unmasker('My name is <mask> <mask>' , top_k=2 )
self.assertEqual(
nested_simplify(_UpperCAmelCase , decimals=6 ) , [
[
{
'score': 2.2e-05,
'token': 35676,
'token_str': ' Maul',
'sequence': '<s>My name is Maul<mask></s>',
},
{'score': 2.2e-05, 'token': 16416, 'token_str': 'ELS', 'sequence': '<s>My name isELS<mask></s>'},
],
[
{
'score': 2.2e-05,
'token': 35676,
'token_str': ' Maul',
'sequence': '<s>My name is<mask> Maul</s>',
},
{'score': 2.2e-05, 'token': 16416, 'token_str': 'ELS', 'sequence': '<s>My name is<mask>ELS</s>'},
],
] , )
@require_torch_gpu
def _UpperCAmelCase ( self ) -> Optional[Any]:
UpperCamelCase_ = pipeline('fill-mask' , model='hf-internal-testing/tiny-random-distilbert' , device=0 , framework='pt' )
# convert model to fp16
pipe.model.half()
UpperCamelCase_ = pipe('Paris is the [MASK] of France.' )
# We actually don't care about the result, we just want to make sure
# it works, meaning the float16 tensor got casted back to float32
# for postprocessing.
self.assertIsInstance(_UpperCAmelCase , _UpperCAmelCase )
@slow
@require_torch
def _UpperCAmelCase ( self ) -> List[Any]:
UpperCamelCase_ = pipeline(task='fill-mask' , model='distilroberta-base' , top_k=2 , framework='pt' )
self.run_large_test(_UpperCAmelCase )
@slow
@require_tf
def _UpperCAmelCase ( self ) -> Union[str, Any]:
UpperCamelCase_ = pipeline(task='fill-mask' , model='distilroberta-base' , top_k=2 , framework='tf' )
self.run_large_test(_UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase ) -> Tuple:
UpperCamelCase_ = unmasker('My name is <mask>' )
self.assertEqual(
nested_simplify(_UpperCAmelCase ) , [
{'sequence': 'My name is John', 'score': 0.0_0_8, 'token': 610, 'token_str': ' John'},
{'sequence': 'My name is Chris', 'score': 0.0_0_7, 'token': 1573, 'token_str': ' Chris'},
] , )
UpperCamelCase_ = unmasker('The largest city in France is <mask>' )
self.assertEqual(
nested_simplify(_UpperCAmelCase ) , [
{
'sequence': 'The largest city in France is Paris',
'score': 0.2_5_1,
'token': 2201,
'token_str': ' Paris',
},
{
'sequence': 'The largest city in France is Lyon',
'score': 0.2_1_4,
'token': 12790,
'token_str': ' Lyon',
},
] , )
UpperCamelCase_ = unmasker('My name is <mask>' , targets=[' Patrick', ' Clara', ' Teven'] , top_k=3 )
self.assertEqual(
nested_simplify(_UpperCAmelCase ) , [
{'sequence': 'My name is Patrick', 'score': 0.0_0_5, 'token': 3499, 'token_str': ' Patrick'},
{'sequence': 'My name is Clara', 'score': 0.0_0_0, 'token': 13606, 'token_str': ' Clara'},
{'sequence': 'My name is Te', 'score': 0.0_0_0, 'token': 2941, 'token_str': ' Te'},
] , )
@require_torch
def _UpperCAmelCase ( self ) -> List[Any]:
UpperCamelCase_ = pipeline(task='fill-mask' , model='sshleifer/tiny-distilroberta-base' , framework='pt' )
UpperCamelCase_ = None
UpperCamelCase_ = None
self.run_pipeline_test(_UpperCAmelCase , [] )
@require_tf
def _UpperCAmelCase ( self ) -> Optional[int]:
UpperCamelCase_ = pipeline(task='fill-mask' , model='sshleifer/tiny-distilroberta-base' , framework='tf' )
UpperCamelCase_ = None
UpperCamelCase_ = None
self.run_pipeline_test(_UpperCAmelCase , [] )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase ) -> Optional[Any]:
if tokenizer is None or tokenizer.mask_token_id is None:
self.skipTest('The provided tokenizer has no mask token, (probably reformer or wav2vec2)' )
UpperCamelCase_ = FillMaskPipeline(model=_UpperCAmelCase , tokenizer=_UpperCAmelCase )
UpperCamelCase_ = [
f"""This is another {tokenizer.mask_token} test""",
]
return fill_masker, examples
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> Union[str, Any]:
UpperCamelCase_ = fill_masker.tokenizer
UpperCamelCase_ = fill_masker.model
UpperCamelCase_ = fill_masker(
f"""This is a {tokenizer.mask_token}""" , )
self.assertEqual(
_UpperCAmelCase , [
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
] , )
UpperCamelCase_ = fill_masker([f"""This is a {tokenizer.mask_token}"""] )
self.assertEqual(
_UpperCAmelCase , [
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
] , )
UpperCamelCase_ = fill_masker([f"""This is a {tokenizer.mask_token}""", f"""Another {tokenizer.mask_token} great test."""] )
self.assertEqual(
_UpperCAmelCase , [
[
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
],
[
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
],
] , )
with self.assertRaises(_UpperCAmelCase ):
fill_masker([None] )
# No mask_token is not supported
with self.assertRaises(_UpperCAmelCase ):
fill_masker('This is' )
self.run_test_top_k(_UpperCAmelCase , _UpperCAmelCase )
self.run_test_targets(_UpperCAmelCase , _UpperCAmelCase )
self.run_test_top_k_targets(_UpperCAmelCase , _UpperCAmelCase )
self.fill_mask_with_duplicate_targets_and_top_k(_UpperCAmelCase , _UpperCAmelCase )
self.fill_mask_with_multiple_masks(_UpperCAmelCase , _UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> Optional[Any]:
UpperCamelCase_ = tokenizer.get_vocab()
UpperCamelCase_ = sorted(vocab.keys() )[:2]
# Pipeline argument
UpperCamelCase_ = FillMaskPipeline(model=_UpperCAmelCase , tokenizer=_UpperCAmelCase , targets=_UpperCAmelCase )
UpperCamelCase_ = fill_masker(f"""This is a {tokenizer.mask_token}""" )
self.assertEqual(
_UpperCAmelCase , [
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
] , )
UpperCamelCase_ = {vocab[el] for el in targets}
self.assertEqual({el['token'] for el in outputs} , _UpperCAmelCase )
UpperCamelCase_ = [tokenizer.decode([x] ) for x in target_ids]
self.assertEqual({el['token_str'] for el in outputs} , set(_UpperCAmelCase ) )
# Call argument
UpperCamelCase_ = FillMaskPipeline(model=_UpperCAmelCase , tokenizer=_UpperCAmelCase )
UpperCamelCase_ = fill_masker(f"""This is a {tokenizer.mask_token}""" , targets=_UpperCAmelCase )
self.assertEqual(
_UpperCAmelCase , [
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
] , )
UpperCamelCase_ = {vocab[el] for el in targets}
self.assertEqual({el['token'] for el in outputs} , _UpperCAmelCase )
UpperCamelCase_ = [tokenizer.decode([x] ) for x in target_ids]
self.assertEqual({el['token_str'] for el in outputs} , set(_UpperCAmelCase ) )
# Score equivalence
UpperCamelCase_ = fill_masker(f"""This is a {tokenizer.mask_token}""" , targets=_UpperCAmelCase )
UpperCamelCase_ = [top_mask['token_str'] for top_mask in outputs]
UpperCamelCase_ = [top_mask['score'] for top_mask in outputs]
# For some BPE tokenizers, `</w>` is removed during decoding, so `token_str` won't be the same as in `targets`.
if set(_UpperCAmelCase ) == set(_UpperCAmelCase ):
UpperCamelCase_ = fill_masker(f"""This is a {tokenizer.mask_token}""" , targets=_UpperCAmelCase )
UpperCamelCase_ = [top_mask['score'] for top_mask in unmasked_targets]
self.assertEqual(nested_simplify(_UpperCAmelCase ) , nested_simplify(_UpperCAmelCase ) )
# Raises with invalid
with self.assertRaises(_UpperCAmelCase ):
UpperCamelCase_ = fill_masker(f"""This is a {tokenizer.mask_token}""" , targets=[] )
# For some tokenizers, `""` is actually in the vocabulary and the expected error won't raised
if "" not in tokenizer.get_vocab():
with self.assertRaises(_UpperCAmelCase ):
UpperCamelCase_ = fill_masker(f"""This is a {tokenizer.mask_token}""" , targets=[''] )
with self.assertRaises(_UpperCAmelCase ):
UpperCamelCase_ = fill_masker(f"""This is a {tokenizer.mask_token}""" , targets='' )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> Tuple:
UpperCamelCase_ = FillMaskPipeline(model=_UpperCAmelCase , tokenizer=_UpperCAmelCase , top_k=2 )
UpperCamelCase_ = fill_masker(f"""This is a {tokenizer.mask_token}""" )
self.assertEqual(
_UpperCAmelCase , [
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
] , )
UpperCamelCase_ = FillMaskPipeline(model=_UpperCAmelCase , tokenizer=_UpperCAmelCase )
UpperCamelCase_ = fill_masker(f"""This is a {tokenizer.mask_token}""" , top_k=2 )
self.assertEqual(
_UpperCAmelCase , [
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
] , )
self.assertEqual(nested_simplify(_UpperCAmelCase ) , nested_simplify(_UpperCAmelCase ) )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> List[Any]:
UpperCamelCase_ = tokenizer.get_vocab()
UpperCamelCase_ = FillMaskPipeline(model=_UpperCAmelCase , tokenizer=_UpperCAmelCase )
# top_k=2, ntargets=3
UpperCamelCase_ = sorted(vocab.keys() )[:3]
UpperCamelCase_ = fill_masker(f"""This is a {tokenizer.mask_token}""" , top_k=2 , targets=_UpperCAmelCase )
# If we use the most probably targets, and filter differently, we should still
# have the same results
UpperCamelCase_ = [el['token_str'] for el in sorted(_UpperCAmelCase , key=lambda _UpperCAmelCase : x["score"] , reverse=_UpperCAmelCase )]
# For some BPE tokenizers, `</w>` is removed during decoding, so `token_str` won't be the same as in `targets`.
if set(_UpperCAmelCase ).issubset(_UpperCAmelCase ):
UpperCamelCase_ = fill_masker(f"""This is a {tokenizer.mask_token}""" , top_k=3 , targets=_UpperCAmelCase )
# They should yield exactly the same result
self.assertEqual(nested_simplify(_UpperCAmelCase ) , nested_simplify(_UpperCAmelCase ) )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> Optional[int]:
UpperCamelCase_ = FillMaskPipeline(model=_UpperCAmelCase , tokenizer=_UpperCAmelCase )
UpperCamelCase_ = tokenizer.get_vocab()
# String duplicates + id duplicates
UpperCamelCase_ = sorted(vocab.keys() )[:3]
UpperCamelCase_ = [targets[0], targets[1], targets[0], targets[2], targets[1]]
UpperCamelCase_ = fill_masker(f"""My name is {tokenizer.mask_token}""" , targets=_UpperCAmelCase , top_k=10 )
# The target list contains duplicates, so we can't output more
# than them
self.assertEqual(len(_UpperCAmelCase ) , 3 )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> List[str]:
UpperCamelCase_ = FillMaskPipeline(model=_UpperCAmelCase , tokenizer=_UpperCAmelCase )
UpperCamelCase_ = fill_masker(
f"""This is a {tokenizer.mask_token} {tokenizer.mask_token} {tokenizer.mask_token}""" , top_k=2 )
self.assertEqual(
_UpperCAmelCase , [
[
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
],
[
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
],
[
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
],
] , )
| 23 | 0 |
from typing import Optional
from urllib.parse import quote
import huggingface_hub as hfh
from packaging import version
def lowerCamelCase__ ( _a , _a , _a = None):
if version.parse(hfh.__version__).release < version.parse("0.11.0").release:
# old versions of hfh don't url-encode the file path
SCREAMING_SNAKE_CASE : Dict = quote(_a)
return hfh.hf_hub_url(_a , _a , repo_type="dataset" , revision=_a) | 25 |
import gc
import unittest
import numpy as np
import torch
from transformers import CLIPTextConfig, CLIPTextModel, CLIPTokenizer
from diffusers import (
AutoencoderKL,
DDIMScheduler,
StableDiffusionSAGPipeline,
UNetaDConditionModel,
)
from diffusers.utils import slow, torch_device
from diffusers.utils.testing_utils import enable_full_determinism, require_torch_gpu
from ..pipeline_params import TEXT_TO_IMAGE_BATCH_PARAMS, TEXT_TO_IMAGE_IMAGE_PARAMS, TEXT_TO_IMAGE_PARAMS
from ..test_pipelines_common import PipelineLatentTesterMixin, PipelineTesterMixin
enable_full_determinism()
class _a ( UpperCAmelCase__ , UpperCAmelCase__ , unittest.TestCase ):
"""simple docstring"""
A_ = StableDiffusionSAGPipeline
A_ = TEXT_TO_IMAGE_PARAMS
A_ = TEXT_TO_IMAGE_BATCH_PARAMS
A_ = TEXT_TO_IMAGE_IMAGE_PARAMS
A_ = TEXT_TO_IMAGE_IMAGE_PARAMS
A_ = False
def _UpperCAmelCase ( self ) -> Optional[Any]:
torch.manual_seed(0 )
UpperCamelCase_ = UNetaDConditionModel(
block_out_channels=(32, 64) , layers_per_block=2 , sample_size=32 , in_channels=4 , out_channels=4 , down_block_types=('DownBlock2D', 'CrossAttnDownBlock2D') , up_block_types=('CrossAttnUpBlock2D', 'UpBlock2D') , cross_attention_dim=32 , )
UpperCamelCase_ = DDIMScheduler(
beta_start=0.0_0_0_8_5 , beta_end=0.0_1_2 , beta_schedule='scaled_linear' , clip_sample=_UpperCAmelCase , set_alpha_to_one=_UpperCAmelCase , )
torch.manual_seed(0 )
UpperCamelCase_ = AutoencoderKL(
block_out_channels=[32, 64] , in_channels=3 , out_channels=3 , down_block_types=['DownEncoderBlock2D', 'DownEncoderBlock2D'] , up_block_types=['UpDecoderBlock2D', 'UpDecoderBlock2D'] , latent_channels=4 , )
torch.manual_seed(0 )
UpperCamelCase_ = CLIPTextConfig(
bos_token_id=0 , eos_token_id=2 , hidden_size=32 , intermediate_size=37 , layer_norm_eps=1e-05 , num_attention_heads=4 , num_hidden_layers=5 , pad_token_id=1 , vocab_size=1000 , )
UpperCamelCase_ = CLIPTextModel(_UpperCAmelCase )
UpperCamelCase_ = CLIPTokenizer.from_pretrained('hf-internal-testing/tiny-random-clip' )
UpperCamelCase_ = {
'unet': unet,
'scheduler': scheduler,
'vae': vae,
'text_encoder': text_encoder,
'tokenizer': tokenizer,
'safety_checker': None,
'feature_extractor': None,
}
return components
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase=0 ) -> List[Any]:
if str(_UpperCAmelCase ).startswith('mps' ):
UpperCamelCase_ = torch.manual_seed(_UpperCAmelCase )
else:
UpperCamelCase_ = torch.Generator(device=_UpperCAmelCase ).manual_seed(_UpperCAmelCase )
UpperCamelCase_ = {
'prompt': '.',
'generator': generator,
'num_inference_steps': 2,
'guidance_scale': 1.0,
'sag_scale': 1.0,
'output_type': 'numpy',
}
return inputs
def _UpperCAmelCase ( self ) -> Tuple:
super().test_inference_batch_single_identical(expected_max_diff=3e-3 )
@slow
@require_torch_gpu
class _a ( unittest.TestCase ):
"""simple docstring"""
def _UpperCAmelCase ( self ) -> Tuple:
# clean up the VRAM after each test
super().tearDown()
gc.collect()
torch.cuda.empty_cache()
def _UpperCAmelCase ( self ) -> str:
UpperCamelCase_ = StableDiffusionSAGPipeline.from_pretrained('CompVis/stable-diffusion-v1-4' )
UpperCamelCase_ = sag_pipe.to(_UpperCAmelCase )
sag_pipe.set_progress_bar_config(disable=_UpperCAmelCase )
UpperCamelCase_ = '.'
UpperCamelCase_ = torch.manual_seed(0 )
UpperCamelCase_ = sag_pipe(
[prompt] , generator=_UpperCAmelCase , guidance_scale=7.5 , sag_scale=1.0 , num_inference_steps=20 , output_type='np' )
UpperCamelCase_ = output.images
UpperCamelCase_ = image[0, -3:, -3:, -1]
assert image.shape == (1, 512, 512, 3)
UpperCamelCase_ = np.array([0.1_5_6_8, 0.1_7_3_8, 0.1_6_9_5, 0.1_6_9_3, 0.1_5_0_7, 0.1_7_0_5, 0.1_5_4_7, 0.1_7_5_1, 0.1_9_4_9] )
assert np.abs(image_slice.flatten() - expected_slice ).max() < 5e-2
def _UpperCAmelCase ( self ) -> List[Any]:
UpperCamelCase_ = StableDiffusionSAGPipeline.from_pretrained('stabilityai/stable-diffusion-2-1-base' )
UpperCamelCase_ = sag_pipe.to(_UpperCAmelCase )
sag_pipe.set_progress_bar_config(disable=_UpperCAmelCase )
UpperCamelCase_ = '.'
UpperCamelCase_ = torch.manual_seed(0 )
UpperCamelCase_ = sag_pipe(
[prompt] , generator=_UpperCAmelCase , guidance_scale=7.5 , sag_scale=1.0 , num_inference_steps=20 , output_type='np' )
UpperCamelCase_ = output.images
UpperCamelCase_ = image[0, -3:, -3:, -1]
assert image.shape == (1, 512, 512, 3)
UpperCamelCase_ = np.array([0.3_4_5_9, 0.2_8_7_6, 0.2_5_3_7, 0.3_0_0_2, 0.2_6_7_1, 0.2_1_6_0, 0.3_0_2_6, 0.2_2_6_2, 0.2_3_7_1] )
assert np.abs(image_slice.flatten() - expected_slice ).max() < 5e-2
def _UpperCAmelCase ( self ) -> Dict:
UpperCamelCase_ = StableDiffusionSAGPipeline.from_pretrained('stabilityai/stable-diffusion-2-1-base' )
UpperCamelCase_ = sag_pipe.to(_UpperCAmelCase )
sag_pipe.set_progress_bar_config(disable=_UpperCAmelCase )
UpperCamelCase_ = '.'
UpperCamelCase_ = torch.manual_seed(0 )
UpperCamelCase_ = sag_pipe(
[prompt] , width=768 , height=512 , generator=_UpperCAmelCase , guidance_scale=7.5 , sag_scale=1.0 , num_inference_steps=20 , output_type='np' , )
UpperCamelCase_ = output.images
assert image.shape == (1, 512, 768, 3)
| 23 | 0 |
'''simple docstring'''
from typing import TYPE_CHECKING
from ...utils import (
OptionalDependencyNotAvailable,
_LazyModule,
is_sentencepiece_available,
is_tokenizers_available,
is_torch_available,
)
__UpperCamelCase = {"configuration_fnet": ["FNET_PRETRAINED_CONFIG_ARCHIVE_MAP", "FNetConfig"]}
try:
if not is_sentencepiece_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
__UpperCamelCase = ["FNetTokenizer"]
try:
if not is_tokenizers_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
__UpperCamelCase = ["FNetTokenizerFast"]
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
__UpperCamelCase = [
"FNET_PRETRAINED_MODEL_ARCHIVE_LIST",
"FNetForMaskedLM",
"FNetForMultipleChoice",
"FNetForNextSentencePrediction",
"FNetForPreTraining",
"FNetForQuestionAnswering",
"FNetForSequenceClassification",
"FNetForTokenClassification",
"FNetLayer",
"FNetModel",
"FNetPreTrainedModel",
]
if TYPE_CHECKING:
from .configuration_fnet import FNET_PRETRAINED_CONFIG_ARCHIVE_MAP, FNetConfig
try:
if not is_sentencepiece_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .tokenization_fnet import FNetTokenizer
try:
if not is_tokenizers_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .tokenization_fnet_fast import FNetTokenizerFast
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_fnet import (
FNET_PRETRAINED_MODEL_ARCHIVE_LIST,
FNetForMaskedLM,
FNetForMultipleChoice,
FNetForNextSentencePrediction,
FNetForPreTraining,
FNetForQuestionAnswering,
FNetForSequenceClassification,
FNetForTokenClassification,
FNetLayer,
FNetModel,
FNetPreTrainedModel,
)
else:
import sys
__UpperCamelCase = _LazyModule(__name__, globals()["__file__"], _import_structure, module_spec=__spec__)
| 26 |
from __future__ import annotations
from sys import maxsize
from typing import Generic, TypeVar
snake_case__ : List[str] = TypeVar("""T""")
def _snake_case (__lowercase):
return (position - 1) // 2
def _snake_case (__lowercase):
return (2 * position) + 1
def _snake_case (__lowercase):
return (2 * position) + 2
class _a ( Generic[T] ):
"""simple docstring"""
def __init__( self ) -> None:
UpperCamelCase_ = []
UpperCamelCase_ = {}
UpperCamelCase_ = 0
def __len__( self ) -> int:
return self.elements
def __repr__( self ) -> str:
return str(self.heap )
def _UpperCAmelCase ( self ) -> bool:
# Check if the priority queue is empty
return self.elements == 0
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> None:
# Add an element with given priority to the queue
self.heap.append((elem, weight) )
UpperCamelCase_ = self.elements
self.elements += 1
self._bubble_up(_UpperCAmelCase )
def _UpperCAmelCase ( self ) -> T:
# Remove and return the element with lowest weight (highest priority)
if self.elements > 1:
self._swap_nodes(0 , self.elements - 1 )
UpperCamelCase_ , UpperCamelCase_ = self.heap.pop()
del self.position_map[elem]
self.elements -= 1
if self.elements > 0:
UpperCamelCase_ , UpperCamelCase_ = self.heap[0]
self._bubble_down(_UpperCAmelCase )
return elem
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> None:
# Update the weight of the given key
UpperCamelCase_ = self.position_map[elem]
UpperCamelCase_ = (elem, weight)
if position > 0:
UpperCamelCase_ = get_parent_position(_UpperCAmelCase )
UpperCamelCase_ , UpperCamelCase_ = self.heap[parent_position]
if parent_weight > weight:
self._bubble_up(_UpperCAmelCase )
else:
self._bubble_down(_UpperCAmelCase )
else:
self._bubble_down(_UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase ) -> None:
# Place a node at the proper position (upward movement) [to be used internally
# only]
UpperCamelCase_ = self.position_map[elem]
if curr_pos == 0:
return None
UpperCamelCase_ = get_parent_position(_UpperCAmelCase )
UpperCamelCase_ , UpperCamelCase_ = self.heap[curr_pos]
UpperCamelCase_ , UpperCamelCase_ = self.heap[parent_position]
if parent_weight > weight:
self._swap_nodes(_UpperCAmelCase , _UpperCAmelCase )
return self._bubble_up(_UpperCAmelCase )
return None
def _UpperCAmelCase ( self , _UpperCAmelCase ) -> None:
# Place a node at the proper position (downward movement) [to be used
# internally only]
UpperCamelCase_ = self.position_map[elem]
UpperCamelCase_ , UpperCamelCase_ = self.heap[curr_pos]
UpperCamelCase_ = get_child_left_position(_UpperCAmelCase )
UpperCamelCase_ = get_child_right_position(_UpperCAmelCase )
if child_left_position < self.elements and child_right_position < self.elements:
UpperCamelCase_ , UpperCamelCase_ = self.heap[child_left_position]
UpperCamelCase_ , UpperCamelCase_ = self.heap[child_right_position]
if child_right_weight < child_left_weight and child_right_weight < weight:
self._swap_nodes(_UpperCAmelCase , _UpperCAmelCase )
return self._bubble_down(_UpperCAmelCase )
if child_left_position < self.elements:
UpperCamelCase_ , UpperCamelCase_ = self.heap[child_left_position]
if child_left_weight < weight:
self._swap_nodes(_UpperCAmelCase , _UpperCAmelCase )
return self._bubble_down(_UpperCAmelCase )
else:
return None
if child_right_position < self.elements:
UpperCamelCase_ , UpperCamelCase_ = self.heap[child_right_position]
if child_right_weight < weight:
self._swap_nodes(_UpperCAmelCase , _UpperCAmelCase )
return self._bubble_down(_UpperCAmelCase )
return None
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> None:
# Swap the nodes at the given positions
UpperCamelCase_ = self.heap[nodea_pos][0]
UpperCamelCase_ = self.heap[nodea_pos][0]
UpperCamelCase_ , UpperCamelCase_ = (
self.heap[nodea_pos],
self.heap[nodea_pos],
)
UpperCamelCase_ = nodea_pos
UpperCamelCase_ = nodea_pos
class _a ( Generic[T] ):
"""simple docstring"""
def __init__( self ) -> None:
UpperCamelCase_ = {}
UpperCamelCase_ = 0
def __repr__( self ) -> str:
return str(self.connections )
def __len__( self ) -> int:
return self.nodes
def _UpperCAmelCase ( self , _UpperCAmelCase ) -> None:
# Add a node in the graph if it is not in the graph
if node not in self.connections:
UpperCamelCase_ = {}
self.nodes += 1
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase ) -> None:
# Add an edge between 2 nodes in the graph
self.add_node(_UpperCAmelCase )
self.add_node(_UpperCAmelCase )
UpperCamelCase_ = weight
UpperCamelCase_ = weight
def _snake_case (__lowercase , ):
UpperCamelCase_ = {node: maxsize for node in graph.connections}
UpperCamelCase_ = {node: None for node in graph.connections}
UpperCamelCase_ = MinPriorityQueue()
for node, weight in dist.items():
priority_queue.push(__lowercase , __lowercase)
if priority_queue.is_empty():
return dist, parent
# initialization
UpperCamelCase_ = priority_queue.extract_min()
UpperCamelCase_ = 0
for neighbour in graph.connections[node]:
if dist[neighbour] > dist[node] + graph.connections[node][neighbour]:
UpperCamelCase_ = dist[node] + graph.connections[node][neighbour]
priority_queue.update_key(__lowercase , dist[neighbour])
UpperCamelCase_ = node
# running prim's algorithm
while not priority_queue.is_empty():
UpperCamelCase_ = priority_queue.extract_min()
for neighbour in graph.connections[node]:
if dist[neighbour] > dist[node] + graph.connections[node][neighbour]:
UpperCamelCase_ = dist[node] + graph.connections[node][neighbour]
priority_queue.update_key(__lowercase , dist[neighbour])
UpperCamelCase_ = node
return dist, parent
| 23 | 0 |
from collections import OrderedDict
from typing import TYPE_CHECKING, Any, Mapping, Optional
from ...configuration_utils import PretrainedConfig
from ...onnx import OnnxConfig
from ...utils import logging
if TYPE_CHECKING:
from ... import FeatureExtractionMixin, TensorType
__A : Optional[int] = logging.get_logger(__name__)
__A : Any = {
"openai/imagegpt-small": "",
"openai/imagegpt-medium": "",
"openai/imagegpt-large": "",
}
class lowerCamelCase( __snake_case ):
'''simple docstring'''
__magic_name__ = 'imagegpt'
__magic_name__ = ['past_key_values']
__magic_name__ = {
'hidden_size': 'n_embd',
'max_position_embeddings': 'n_positions',
'num_attention_heads': 'n_head',
'num_hidden_layers': 'n_layer',
}
def __init__( self , snake_case_=512 + 1 , snake_case_=32 * 32 , snake_case_=512 , snake_case_=24 , snake_case_=8 , snake_case_=None , snake_case_="quick_gelu" , snake_case_=0.1 , snake_case_=0.1 , snake_case_=0.1 , snake_case_=1E-5 , snake_case_=0.02 , snake_case_=True , snake_case_=True , snake_case_=False , snake_case_=False , snake_case_=False , **snake_case_ , ):
_A = vocab_size
_A = n_positions
_A = n_embd
_A = n_layer
_A = n_head
_A = n_inner
_A = activation_function
_A = resid_pdrop
_A = embd_pdrop
_A = attn_pdrop
_A = layer_norm_epsilon
_A = initializer_range
_A = scale_attn_weights
_A = use_cache
_A = scale_attn_by_inverse_layer_idx
_A = reorder_and_upcast_attn
_A = tie_word_embeddings
super().__init__(tie_word_embeddings=snake_case_ , **snake_case_ )
class lowerCamelCase( __snake_case ):
'''simple docstring'''
@property
def lowerCAmelCase__ ( self ):
return OrderedDict(
[
('input_ids', {0: 'batch', 1: 'sequence'}),
] )
def lowerCAmelCase__ ( self , snake_case_ , snake_case_ = 1 , snake_case_ = -1 , snake_case_ = False , snake_case_ = None , snake_case_ = 3 , snake_case_ = 32 , snake_case_ = 32 , ):
_A = self._generate_dummy_images(snake_case_ , snake_case_ , snake_case_ , snake_case_ )
_A = dict(preprocessor(images=snake_case_ , return_tensors=snake_case_ ) )
return inputs
| 27 |
from __future__ import annotations
import sys
from collections import deque
from typing import Generic, TypeVar
snake_case__ : Dict = TypeVar("""T""")
class _a ( Generic[T] ):
"""simple docstring"""
A_ = 42 # Cache store of keys
A_ = 42 # References of the keys in cache
A_ = 10 # Maximum capacity of cache
def __init__( self , _UpperCAmelCase ) -> None:
UpperCamelCase_ = deque()
UpperCamelCase_ = set()
if not n:
UpperCamelCase_ = sys.maxsize
elif n < 0:
raise ValueError('n should be an integer greater than 0.' )
else:
UpperCamelCase_ = n
def _UpperCAmelCase ( self , _UpperCAmelCase ) -> None:
if x not in self.key_reference:
if len(self.dq_store ) == LRUCache._MAX_CAPACITY:
UpperCamelCase_ = self.dq_store.pop()
self.key_reference.remove(_UpperCAmelCase )
else:
self.dq_store.remove(_UpperCAmelCase )
self.dq_store.appendleft(_UpperCAmelCase )
self.key_reference.add(_UpperCAmelCase )
def _UpperCAmelCase ( self ) -> None:
for k in self.dq_store:
print(_UpperCAmelCase )
def __repr__( self ) -> str:
return f"""LRUCache({self._MAX_CAPACITY}) => {list(self.dq_store )}"""
if __name__ == "__main__":
import doctest
doctest.testmod()
snake_case__ : LRUCache[str | int] = LRUCache(4)
lru_cache.refer("""A""")
lru_cache.refer(2)
lru_cache.refer(3)
lru_cache.refer("""A""")
lru_cache.refer(4)
lru_cache.refer(5)
lru_cache.display()
print(lru_cache)
assert str(lru_cache) == "LRUCache(4) => [5, 4, 'A', 3]"
| 23 | 0 |
'''simple docstring'''
from abc import ABC, abstractmethod
from typing import List, Optional
class _a ( SCREAMING_SNAKE_CASE ):
'''simple docstring'''
def __init__( self ):
'''simple docstring'''
self.test()
def UpperCamelCase_ ( self ):
'''simple docstring'''
SCREAMING_SNAKE_CASE : Optional[int] = 0
SCREAMING_SNAKE_CASE : Optional[int] = False
while not completed:
if counter == 1:
self.reset()
SCREAMING_SNAKE_CASE : str = self.advance()
if not self.does_advance(A ):
raise Exception(
'Custom Constraint is not defined correctly. self.does_advance(self.advance()) must be true.' )
SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE : Union[str, Any] = self.update(A )
counter += 1
if counter > 10_000:
raise Exception('update() does not fulfill the constraint.' )
if self.remaining() != 0:
raise Exception('Custom Constraint is not defined correctly.' )
@abstractmethod
def UpperCamelCase_ ( self ):
'''simple docstring'''
raise NotImplementedError(
F"{self.__class__} is an abstract class. Only classes inheriting this class can be called." )
@abstractmethod
def UpperCamelCase_ ( self, A ):
'''simple docstring'''
raise NotImplementedError(
F"{self.__class__} is an abstract class. Only classes inheriting this class can be called." )
@abstractmethod
def UpperCamelCase_ ( self, A ):
'''simple docstring'''
raise NotImplementedError(
F"{self.__class__} is an abstract class. Only classes inheriting this class can be called." )
@abstractmethod
def UpperCamelCase_ ( self ):
'''simple docstring'''
raise NotImplementedError(
F"{self.__class__} is an abstract class. Only classes inheriting this class can be called." )
@abstractmethod
def UpperCamelCase_ ( self ):
'''simple docstring'''
raise NotImplementedError(
F"{self.__class__} is an abstract class. Only classes inheriting this class can be called." )
@abstractmethod
def UpperCamelCase_ ( self, A=False ):
'''simple docstring'''
raise NotImplementedError(
F"{self.__class__} is an abstract class. Only classes inheriting this class can be called." )
class _a ( SCREAMING_SNAKE_CASE ):
'''simple docstring'''
def __init__( self, A ):
'''simple docstring'''
super(A, self ).__init__()
if not isinstance(A, A ) or len(A ) == 0:
raise ValueError(F"`token_ids` has to be a non-empty list, but is {token_ids}." )
if any((not isinstance(A, A ) or token_id < 0) for token_id in token_ids ):
raise ValueError(F"Each list in `token_ids` has to be a list of positive integers, but is {token_ids}." )
SCREAMING_SNAKE_CASE : Tuple = token_ids
SCREAMING_SNAKE_CASE : Tuple = len(self.token_ids )
SCREAMING_SNAKE_CASE : Optional[int] = -1 # the index of the currently fulfilled step
SCREAMING_SNAKE_CASE : Any = False
def UpperCamelCase_ ( self ):
'''simple docstring'''
if self.completed:
return None
return self.token_ids[self.fulfilled_idx + 1]
def UpperCamelCase_ ( self, A ):
'''simple docstring'''
if not isinstance(A, A ):
raise ValueError(F"`token_id` has to be an `int`, but is {token_id} of type {type(A )}" )
if self.completed:
return False
return token_id == self.token_ids[self.fulfilled_idx + 1]
def UpperCamelCase_ ( self, A ):
'''simple docstring'''
if not isinstance(A, A ):
raise ValueError(F"`token_id` has to be an `int`, but is {token_id} of type {type(A )}" )
SCREAMING_SNAKE_CASE : Dict = False
SCREAMING_SNAKE_CASE : str = False
SCREAMING_SNAKE_CASE : int = False
if self.does_advance(A ):
self.fulfilled_idx += 1
SCREAMING_SNAKE_CASE : int = True
if self.fulfilled_idx == (self.seqlen - 1):
SCREAMING_SNAKE_CASE : Optional[int] = True
SCREAMING_SNAKE_CASE : Union[str, Any] = completed
else:
# failed to make progress.
SCREAMING_SNAKE_CASE : int = True
self.reset()
return stepped, completed, reset
def UpperCamelCase_ ( self ):
'''simple docstring'''
SCREAMING_SNAKE_CASE : Optional[Any] = False
SCREAMING_SNAKE_CASE : Dict = 0
def UpperCamelCase_ ( self ):
'''simple docstring'''
return self.seqlen - (self.fulfilled_idx + 1)
def UpperCamelCase_ ( self, A=False ):
'''simple docstring'''
SCREAMING_SNAKE_CASE : str = PhrasalConstraint(self.token_ids )
if stateful:
SCREAMING_SNAKE_CASE : str = self.seqlen
SCREAMING_SNAKE_CASE : List[Any] = self.fulfilled_idx
SCREAMING_SNAKE_CASE : List[Any] = self.completed
return new_constraint
class _a :
'''simple docstring'''
def __init__( self, A, A=True ):
'''simple docstring'''
SCREAMING_SNAKE_CASE : Any = max([len(A ) for one in nested_token_ids] )
SCREAMING_SNAKE_CASE : Optional[int] = {}
for token_ids in nested_token_ids:
SCREAMING_SNAKE_CASE : Dict = root
for tidx, token_id in enumerate(A ):
if token_id not in level:
SCREAMING_SNAKE_CASE : Dict = {}
SCREAMING_SNAKE_CASE : Optional[Any] = level[token_id]
if no_subsets and self.has_subsets(A, A ):
raise ValueError(
'Each list in `nested_token_ids` can\'t be a complete subset of another list, but is'
F" {nested_token_ids}." )
SCREAMING_SNAKE_CASE : Any = root
def UpperCamelCase_ ( self, A ):
'''simple docstring'''
SCREAMING_SNAKE_CASE : List[Any] = self.trie
for current_token in current_seq:
SCREAMING_SNAKE_CASE : Optional[Any] = start[current_token]
SCREAMING_SNAKE_CASE : Optional[Any] = list(start.keys() )
return next_tokens
def UpperCamelCase_ ( self, A ):
'''simple docstring'''
SCREAMING_SNAKE_CASE : List[str] = self.next_tokens(A )
return len(A ) == 0
def UpperCamelCase_ ( self, A ):
'''simple docstring'''
SCREAMING_SNAKE_CASE : List[Any] = list(root.values() )
if len(A ) == 0:
return 1
else:
return sum([self.count_leaves(A ) for nn in next_nodes] )
def UpperCamelCase_ ( self, A, A ):
'''simple docstring'''
SCREAMING_SNAKE_CASE : str = self.count_leaves(A )
return len(A ) != leaf_count
class _a ( SCREAMING_SNAKE_CASE ):
'''simple docstring'''
def __init__( self, A ):
'''simple docstring'''
super(A, self ).__init__()
if not isinstance(A, A ) or len(A ) == 0:
raise ValueError(F"`nested_token_ids` has to be a non-empty list, but is {nested_token_ids}." )
if any(not isinstance(A, A ) for token_ids in nested_token_ids ):
raise ValueError(F"`nested_token_ids` has to be a list of lists, but is {nested_token_ids}." )
if any(
any((not isinstance(A, A ) or token_id < 0) for token_id in token_ids )
for token_ids in nested_token_ids ):
raise ValueError(
F"Each list in `nested_token_ids` has to be a list of positive integers, but is {nested_token_ids}." )
SCREAMING_SNAKE_CASE : Dict = DisjunctiveTrie(A )
SCREAMING_SNAKE_CASE : int = nested_token_ids
SCREAMING_SNAKE_CASE : int = self.trie.max_height
SCREAMING_SNAKE_CASE : str = []
SCREAMING_SNAKE_CASE : List[str] = False
def UpperCamelCase_ ( self ):
'''simple docstring'''
SCREAMING_SNAKE_CASE : str = self.trie.next_tokens(self.current_seq )
if len(A ) == 0:
return None
else:
return token_list
def UpperCamelCase_ ( self, A ):
'''simple docstring'''
if not isinstance(A, A ):
raise ValueError(F"`token_id` is supposed to be type `int`, but is {token_id} of type {type(A )}" )
SCREAMING_SNAKE_CASE : List[str] = self.trie.next_tokens(self.current_seq )
return token_id in next_tokens
def UpperCamelCase_ ( self, A ):
'''simple docstring'''
if not isinstance(A, A ):
raise ValueError(F"`token_id` is supposed to be type `int`, but is {token_id} of type {type(A )}" )
SCREAMING_SNAKE_CASE : Optional[Any] = False
SCREAMING_SNAKE_CASE : Optional[int] = False
SCREAMING_SNAKE_CASE : Optional[Any] = False
if self.does_advance(A ):
self.current_seq.append(A )
SCREAMING_SNAKE_CASE : Tuple = True
else:
SCREAMING_SNAKE_CASE : Dict = True
self.reset()
SCREAMING_SNAKE_CASE : int = self.trie.reached_leaf(self.current_seq )
SCREAMING_SNAKE_CASE : List[str] = completed
return stepped, completed, reset
def UpperCamelCase_ ( self ):
'''simple docstring'''
SCREAMING_SNAKE_CASE : List[str] = False
SCREAMING_SNAKE_CASE : Optional[int] = []
def UpperCamelCase_ ( self ):
'''simple docstring'''
if self.completed:
# since this can be completed without reaching max height
return 0
else:
return self.seqlen - len(self.current_seq )
def UpperCamelCase_ ( self, A=False ):
'''simple docstring'''
SCREAMING_SNAKE_CASE : str = DisjunctiveConstraint(self.token_ids )
if stateful:
SCREAMING_SNAKE_CASE : Tuple = self.seqlen
SCREAMING_SNAKE_CASE : Dict = self.current_seq
SCREAMING_SNAKE_CASE : str = self.completed
return new_constraint
class _a :
'''simple docstring'''
def __init__( self, A ):
'''simple docstring'''
SCREAMING_SNAKE_CASE : str = constraints
# max # of steps required to fulfill a given constraint
SCREAMING_SNAKE_CASE : List[str] = max([c.seqlen for c in constraints] )
SCREAMING_SNAKE_CASE : str = len(A )
SCREAMING_SNAKE_CASE : Any = False
self.init_state()
def UpperCamelCase_ ( self ):
'''simple docstring'''
SCREAMING_SNAKE_CASE : List[Any] = []
SCREAMING_SNAKE_CASE : Any = None
SCREAMING_SNAKE_CASE : str = [constraint.copy(stateful=A ) for constraint in self.constraints]
def UpperCamelCase_ ( self ):
'''simple docstring'''
SCREAMING_SNAKE_CASE : List[Any] = 0
if self.inprogress_constraint:
# extra points for having a constraint mid-fulfilled
add += self.max_seqlen - self.inprogress_constraint.remaining()
return (len(self.complete_constraints ) * self.max_seqlen) + add
def UpperCamelCase_ ( self ):
'''simple docstring'''
SCREAMING_SNAKE_CASE : str = []
if self.inprogress_constraint is None:
for constraint in self.pending_constraints: # "pending" == "unfulfilled yet"
SCREAMING_SNAKE_CASE : List[str] = constraint.advance()
if isinstance(A, A ):
token_list.append(A )
elif isinstance(A, A ):
token_list.extend(A )
else:
SCREAMING_SNAKE_CASE : List[Any] = self.inprogress_constraint.advance()
if isinstance(A, A ):
token_list.append(A )
elif isinstance(A, A ):
token_list.extend(A )
if len(A ) == 0:
return None
else:
return token_list
def UpperCamelCase_ ( self, A ):
'''simple docstring'''
self.init_state()
if token_ids is not None:
for token in token_ids:
# completes or steps **one** constraint
SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE : Union[str, Any] = self.add(A )
# the entire list of constraints are fulfilled
if self.completed:
break
def UpperCamelCase_ ( self, A ):
'''simple docstring'''
if not isinstance(A, A ):
raise ValueError(F"`token_id` should be an `int`, but is `{token_id}`." )
SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE : List[Any] = False, False
if self.completed:
SCREAMING_SNAKE_CASE : Tuple = True
SCREAMING_SNAKE_CASE : Dict = False
return complete, stepped
if self.inprogress_constraint is not None:
# In the middle of fulfilling a constraint. If the `token_id` *does* makes an incremental progress to current
# job, simply update the state
SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE : Optional[int] = self.inprogress_constraint.update(A )
if reset:
# 1. If the next token breaks the progress, then we must restart.
# e.g. constraint = "I love pies" and sequence so far is "I love" but `token_id` == "books".
# But that doesn't mean we self.init_state(), since we only reset the state for this particular
# constraint, not the full list of constraints.
self.pending_constraints.append(self.inprogress_constraint.copy(stateful=A ) )
SCREAMING_SNAKE_CASE : Dict = None
if complete:
# 2. If the next token completes the constraint, move it to completed list, set
# inprogress to None. If there are no pending constraints either, then this full list of constraints
# is complete.
self.complete_constraints.append(self.inprogress_constraint )
SCREAMING_SNAKE_CASE : Any = None
if len(self.pending_constraints ) == 0:
# we're done!
SCREAMING_SNAKE_CASE : Optional[int] = True
else:
# Not in the middle of fulfilling a constraint. So does this `token_id` helps us step towards any of our list
# of constraints?
for cidx, pending_constraint in enumerate(self.pending_constraints ):
if pending_constraint.does_advance(A ):
SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE : Dict = pending_constraint.update(A )
if not stepped:
raise Exception(
'`constraint.update(token_id)` is not yielding incremental progress, '
'even though `constraint.does_advance(token_id)` is true.' )
if complete:
self.complete_constraints.append(A )
SCREAMING_SNAKE_CASE : Optional[Any] = None
if not complete and stepped:
SCREAMING_SNAKE_CASE : Union[str, Any] = pending_constraint
if complete or stepped:
# If we made any progress at all, then it's at least not a "pending constraint".
SCREAMING_SNAKE_CASE : Any = (
self.pending_constraints[:cidx] + self.pending_constraints[cidx + 1 :]
)
if len(self.pending_constraints ) == 0 and self.inprogress_constraint is None:
# If there's no longer any pending after this and no inprogress either, then we must be
# complete.
SCREAMING_SNAKE_CASE : Dict = True
break # prevent accidentally stepping through multiple constraints with just one token.
return complete, stepped
def UpperCamelCase_ ( self, A=True ):
'''simple docstring'''
SCREAMING_SNAKE_CASE : List[str] = ConstraintListState(self.constraints ) # we actually never though self.constraints objects
# throughout this process. So it's at initialization state.
if stateful:
SCREAMING_SNAKE_CASE : int = [
constraint.copy(stateful=A ) for constraint in self.complete_constraints
]
if self.inprogress_constraint is not None:
SCREAMING_SNAKE_CASE : List[str] = self.inprogress_constraint.copy(stateful=A )
SCREAMING_SNAKE_CASE : Optional[Any] = [constraint.copy() for constraint in self.pending_constraints]
return new_state
| 28 |
import numpy as np
def _snake_case (__lowercase):
return 1 / (1 + np.exp(-vector))
def _snake_case (__lowercase):
return vector * sigmoid(__lowercase)
if __name__ == "__main__":
import doctest
doctest.testmod()
| 23 | 0 |
"""simple docstring"""
# This is the module that test_patching.py uses to test patch_submodule()
import os # noqa: this is just for tests
import os as renamed_os # noqa: this is just for tests
from os import path # noqa: this is just for tests
from os import path as renamed_path # noqa: this is just for tests
from os.path import join # noqa: this is just for tests
from os.path import join as renamed_join # noqa: this is just for tests
A_ = open # noqa: we just need to have a builtin inside this module to test it properly
| 29 |
import math
from datetime import datetime, timedelta
def _snake_case (__lowercase):
UpperCamelCase_ = year % 19
UpperCamelCase_ = year % 4
UpperCamelCase_ = year % 7
UpperCamelCase_ = math.floor(year / 100)
UpperCamelCase_ = math.floor((13 + 8 * leap_day_inhibits) / 25)
UpperCamelCase_ = leap_day_inhibits / 4
UpperCamelCase_ = (
15 - lunar_orbit_correction + leap_day_inhibits - leap_day_reinstall_number
) % 30
UpperCamelCase_ = (4 + leap_day_inhibits - leap_day_reinstall_number) % 7
# days to be added to March 21
UpperCamelCase_ = (19 * metonic_cycle + secular_moon_shift) % 30
# PHM -> Paschal Full Moon
UpperCamelCase_ = (
2 * julian_leap_year
+ 4 * non_leap_year
+ 6 * days_to_add
+ century_starting_point
) % 7
if days_to_add == 29 and days_from_phm_to_sunday == 6:
return datetime(__lowercase , 4 , 19)
elif days_to_add == 28 and days_from_phm_to_sunday == 6:
return datetime(__lowercase , 4 , 18)
else:
return datetime(__lowercase , 3 , 22) + timedelta(
days=int(days_to_add + days_from_phm_to_sunday))
if __name__ == "__main__":
for year in (1_9_9_4, 2_0_0_0, 2_0_1_0, 2_0_2_1, 2_0_2_3):
snake_case__ : Dict = """will be""" if year > datetime.now().year else """was"""
print(f'Easter in {year} {tense} {gauss_easter(year)}')
| 23 | 0 |
from transformers import BertTokenizer, EncoderDecoderModel, SeqaSeqTrainer, SeqaSeqTrainingArguments
from transformers.testing_utils import TestCasePlus, require_torch, slow
from transformers.utils import is_datasets_available
if is_datasets_available():
import datasets
class __a( _a ):
"""simple docstring"""
@slow
@require_torch
def a__ ( self ) -> Tuple:
UpperCAmelCase_ : Tuple = EncoderDecoderModel.from_encoder_decoder_pretrained('''prajjwal1/bert-tiny''' ,'''prajjwal1/bert-tiny''' )
UpperCAmelCase_ : Union[str, Any] = BertTokenizer.from_pretrained('''bert-base-uncased''' )
UpperCAmelCase_ : Optional[int] = bertabert.config.encoder.vocab_size
UpperCAmelCase_ : Any = tokenizer.sep_token_id
UpperCAmelCase_ : Optional[int] = tokenizer.cls_token_id
UpperCAmelCase_ : List[Any] = 128
UpperCAmelCase_ : int = datasets.load_dataset('''cnn_dailymail''' ,'''3.0.0''' ,split='''train[:1%]''' )
UpperCAmelCase_ : str = datasets.load_dataset('''cnn_dailymail''' ,'''3.0.0''' ,split='''validation[:1%]''' )
UpperCAmelCase_ : str = train_dataset.select(range(32 ) )
UpperCAmelCase_ : int = val_dataset.select(range(16 ) )
UpperCAmelCase_ : Optional[Any] = 4
def _map_to_encoder_decoder_inputs(_SCREAMING_SNAKE_CASE ):
# Tokenizer will automatically set [BOS] <text> [EOS]
UpperCAmelCase_ : List[str] = tokenizer(batch['''article'''] ,padding='''max_length''' ,truncation=_SCREAMING_SNAKE_CASE ,max_length=512 )
UpperCAmelCase_ : str = tokenizer(batch['''highlights'''] ,padding='''max_length''' ,truncation=_SCREAMING_SNAKE_CASE ,max_length=128 )
UpperCAmelCase_ : List[str] = inputs.input_ids
UpperCAmelCase_ : Optional[Any] = inputs.attention_mask
UpperCAmelCase_ : Optional[Any] = outputs.input_ids
UpperCAmelCase_ : Optional[int] = outputs.input_ids.copy()
UpperCAmelCase_ : Any = [
[-100 if token == tokenizer.pad_token_id else token for token in labels] for labels in batch['''labels''']
]
UpperCAmelCase_ : Dict = outputs.attention_mask
assert all(len(_SCREAMING_SNAKE_CASE ) == 512 for x in inputs.input_ids )
assert all(len(_SCREAMING_SNAKE_CASE ) == 128 for x in outputs.input_ids )
return batch
def _compute_metrics(_SCREAMING_SNAKE_CASE ):
UpperCAmelCase_ : List[str] = pred.label_ids
UpperCAmelCase_ : Dict = pred.predictions
# all unnecessary tokens are removed
UpperCAmelCase_ : List[str] = tokenizer.batch_decode(_SCREAMING_SNAKE_CASE ,skip_special_tokens=_SCREAMING_SNAKE_CASE )
UpperCAmelCase_ : List[str] = tokenizer.batch_decode(_SCREAMING_SNAKE_CASE ,skip_special_tokens=_SCREAMING_SNAKE_CASE )
UpperCAmelCase_ : str = sum([int(pred_str[i] == label_str[i] ) for i in range(len(_SCREAMING_SNAKE_CASE ) )] ) / len(_SCREAMING_SNAKE_CASE )
return {"accuracy": accuracy}
# map train dataset
UpperCAmelCase_ : Union[str, Any] = train_dataset.map(
_map_to_encoder_decoder_inputs ,batched=_SCREAMING_SNAKE_CASE ,batch_size=_SCREAMING_SNAKE_CASE ,remove_columns=['''article''', '''highlights'''] ,)
train_dataset.set_format(
type='''torch''' ,columns=['''input_ids''', '''attention_mask''', '''decoder_input_ids''', '''decoder_attention_mask''', '''labels'''] ,)
# same for validation dataset
UpperCAmelCase_ : List[str] = val_dataset.map(
_map_to_encoder_decoder_inputs ,batched=_SCREAMING_SNAKE_CASE ,batch_size=_SCREAMING_SNAKE_CASE ,remove_columns=['''article''', '''highlights'''] ,)
val_dataset.set_format(
type='''torch''' ,columns=['''input_ids''', '''attention_mask''', '''decoder_input_ids''', '''decoder_attention_mask''', '''labels'''] ,)
UpperCAmelCase_ : Optional[int] = self.get_auto_remove_tmp_dir()
UpperCAmelCase_ : List[Any] = SeqaSeqTrainingArguments(
output_dir=_SCREAMING_SNAKE_CASE ,per_device_train_batch_size=_SCREAMING_SNAKE_CASE ,per_device_eval_batch_size=_SCREAMING_SNAKE_CASE ,predict_with_generate=_SCREAMING_SNAKE_CASE ,evaluation_strategy='''steps''' ,do_train=_SCREAMING_SNAKE_CASE ,do_eval=_SCREAMING_SNAKE_CASE ,warmup_steps=0 ,eval_steps=2 ,logging_steps=2 ,)
# instantiate trainer
UpperCAmelCase_ : int = SeqaSeqTrainer(
model=_SCREAMING_SNAKE_CASE ,args=_SCREAMING_SNAKE_CASE ,compute_metrics=_compute_metrics ,train_dataset=_SCREAMING_SNAKE_CASE ,eval_dataset=_SCREAMING_SNAKE_CASE ,tokenizer=_SCREAMING_SNAKE_CASE ,)
# start training
trainer.train() | 30 |
import requests
def _snake_case (__lowercase , __lowercase):
UpperCamelCase_ = {'Content-Type': 'application/json'}
UpperCamelCase_ = requests.post(__lowercase , json={'text': message_body} , headers=__lowercase)
if response.status_code != 200:
UpperCamelCase_ = (
'Request to slack returned an error '
f"""{response.status_code}, the response is:\n{response.text}"""
)
raise ValueError(__lowercase)
if __name__ == "__main__":
# Set the slack url to the one provided by Slack when you create the webhook at
# https://my.slack.com/services/new/incoming-webhook/
send_slack_message("""<YOUR MESSAGE BODY>""", """<SLACK CHANNEL URL>""")
| 23 | 0 |
from unittest.mock import Mock, patch
from file_transfer.send_file import send_file
@patch('socket.socket' )
@patch('builtins.open' )
def UpperCAmelCase_ ( __UpperCAmelCase : int , __UpperCAmelCase : int ) -> Union[str, Any]:
# ===== initialization =====
SCREAMING_SNAKE_CASE_ = Mock()
SCREAMING_SNAKE_CASE_ = conn, Mock()
SCREAMING_SNAKE_CASE_ = iter([1, None] )
SCREAMING_SNAKE_CASE_ = lambda __UpperCAmelCase : next(__UpperCAmelCase )
# ===== invoke =====
send_file(filename='mytext.txt' , testing=__UpperCAmelCase )
# ===== ensurance =====
sock.assert_called_once()
sock.return_value.bind.assert_called_once()
sock.return_value.listen.assert_called_once()
sock.return_value.accept.assert_called_once()
conn.recv.assert_called_once()
file.return_value.__enter__.assert_called_once()
file.return_value.__enter__.return_value.read.assert_called()
conn.send.assert_called_once()
conn.close.assert_called_once()
sock.return_value.shutdown.assert_called_once()
sock.return_value.close.assert_called_once() | 31 |
import re
from pathlib import Path
from unittest import TestCase
import pytest
@pytest.mark.integration
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
def _UpperCAmelCase ( self , _UpperCAmelCase ) -> Dict:
with open(_UpperCAmelCase , encoding='utf-8' ) as input_file:
UpperCamelCase_ = re.compile(R'(?!.*\b(?:encoding|rb|w|wb|w+|wb+|ab|ab+)\b)(?<=\s)(open)\((.*)\)' )
UpperCamelCase_ = input_file.read()
UpperCamelCase_ = regexp.search(_UpperCAmelCase )
return match
def _UpperCAmelCase ( self , _UpperCAmelCase ) -> Dict:
with open(_UpperCAmelCase , encoding='utf-8' ) as input_file:
UpperCamelCase_ = re.compile(R'#[^\r\n]*print\(|\"[^\r\n]*print\(|\"\"\".*?print\(.*?\"\"\"|(print\()' , re.DOTALL )
UpperCamelCase_ = input_file.read()
# use `re.finditer` to handle the case where the ignored groups would be matched first by `re.search`
UpperCamelCase_ = regexp.finditer(_UpperCAmelCase )
UpperCamelCase_ = [match for match in matches if match is not None and match.group(1 ) is not None]
return matches[0] if matches else None
def _UpperCAmelCase ( self ) -> List[str]:
UpperCamelCase_ = Path('./datasets' )
UpperCamelCase_ = list(dataset_paths.absolute().glob('**/*.py' ) )
for dataset in dataset_files:
if self._no_encoding_on_file_open(str(_UpperCAmelCase ) ):
raise AssertionError(f"""open(...) must use utf-8 encoding in {dataset}""" )
def _UpperCAmelCase ( self ) -> Optional[int]:
UpperCamelCase_ = Path('./datasets' )
UpperCamelCase_ = list(dataset_paths.absolute().glob('**/*.py' ) )
for dataset in dataset_files:
if self._no_print_statements(str(_UpperCAmelCase ) ):
raise AssertionError(f"""print statement found in {dataset}. Use datasets.logger/logging instead.""" )
| 23 | 0 |
UpperCAmelCase_ = "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/"
def A__ ( SCREAMING_SNAKE_CASE_ : bytes ) -> bytes:
"""simple docstring"""
if not isinstance(SCREAMING_SNAKE_CASE_ , SCREAMING_SNAKE_CASE_ ):
_UpperCAmelCase = F'''a bytes-like object is required, not \'{data.__class__.__name__}\''''
raise TypeError(SCREAMING_SNAKE_CASE_ )
_UpperCAmelCase = ''''''.join(bin(SCREAMING_SNAKE_CASE_ )[2:].zfill(8 ) for byte in data )
_UpperCAmelCase = len(SCREAMING_SNAKE_CASE_ ) % 6 != 0
if padding_needed:
# The padding that will be added later
_UpperCAmelCase = B'''=''' * ((6 - len(SCREAMING_SNAKE_CASE_ ) % 6) // 2)
# Append binary_stream with arbitrary binary digits (0's by default) to make its
# length a multiple of 6.
binary_stream += "0" * (6 - len(SCREAMING_SNAKE_CASE_ ) % 6)
else:
_UpperCAmelCase = B''''''
# Encode every 6 binary digits to their corresponding Base64 character
return (
"".join(
B64_CHARSET[int(binary_stream[index : index + 6] , 2 )]
for index in range(0 , len(SCREAMING_SNAKE_CASE_ ) , 6 ) ).encode()
+ padding
)
def A__ ( SCREAMING_SNAKE_CASE_ : str ) -> bytes:
"""simple docstring"""
if not isinstance(SCREAMING_SNAKE_CASE_ , SCREAMING_SNAKE_CASE_ ) and not isinstance(SCREAMING_SNAKE_CASE_ , SCREAMING_SNAKE_CASE_ ):
_UpperCAmelCase = (
'''argument should be a bytes-like object or ASCII string, '''
F'''not \'{encoded_data.__class__.__name__}\''''
)
raise TypeError(SCREAMING_SNAKE_CASE_ )
# In case encoded_data is a bytes-like object, make sure it contains only
# ASCII characters so we convert it to a string object
if isinstance(SCREAMING_SNAKE_CASE_ , SCREAMING_SNAKE_CASE_ ):
try:
_UpperCAmelCase = encoded_data.decode('''utf-8''' )
except UnicodeDecodeError:
raise ValueError('''base64 encoded data should only contain ASCII characters''' )
_UpperCAmelCase = encoded_data.count('''=''' )
# Check if the encoded string contains non base64 characters
if padding:
assert all(
char in B64_CHARSET for char in encoded_data[:-padding] ), "Invalid base64 character(s) found."
else:
assert all(
char in B64_CHARSET for char in encoded_data ), "Invalid base64 character(s) found."
# Check the padding
assert len(SCREAMING_SNAKE_CASE_ ) % 4 == 0 and padding < 3, "Incorrect padding"
if padding:
# Remove padding if there is one
_UpperCAmelCase = encoded_data[:-padding]
_UpperCAmelCase = ''''''.join(
bin(B64_CHARSET.index(SCREAMING_SNAKE_CASE_ ) )[2:].zfill(6 ) for char in encoded_data )[: -padding * 2]
else:
_UpperCAmelCase = ''''''.join(
bin(B64_CHARSET.index(SCREAMING_SNAKE_CASE_ ) )[2:].zfill(6 ) for char in encoded_data )
_UpperCAmelCase = [
int(binary_stream[index : index + 8] , 2 )
for index in range(0 , len(SCREAMING_SNAKE_CASE_ ) , 8 )
]
return bytes(SCREAMING_SNAKE_CASE_ )
if __name__ == "__main__":
import doctest
doctest.testmod() | 32 |
import argparse
import random
import joblib
import numpy as np
import torch
from igf.igf import (
SecondaryLearner,
collect_objective_set,
compute_perplexity,
generate_datasets,
load_gpta,
recopy_gpta,
set_seed,
train_secondary_learner,
)
from torch.utils.data import DataLoader, RandomSampler
from transformers import GPTaLMHeadModel
def _snake_case (__lowercase=32 , __lowercase=10 , __lowercase=100 , __lowercase=1026 , __lowercase=True , __lowercase="data/tokenized_stories_train_wikitext103.jbl" , __lowercase="igf_context_pairs.jbl" , ):
set_seed(3)
# generate train_data and objective_set
UpperCamelCase_ , UpperCamelCase_ = generate_datasets(
__lowercase , __lowercase , number=__lowercase , min_len=1026 , trim=__lowercase)
# keeps model same across runs
set_seed(4)
# model, lm_optimizer, lm_scheduler = recopy_gpt2(model, device, max_steps) # store original model weights
# can we train on GPU?
UpperCamelCase_ = torch.device('cuda:0' if torch.cuda.is_available() else 'cpu')
# load pretrained model
UpperCamelCase_ = load_gpta('gpt2').to(__lowercase)
print('computing perplexity on objective set')
UpperCamelCase_ = compute_perplexity(__lowercase , __lowercase , __lowercase).item()
print('perplexity on objective set:' , __lowercase)
# collect igf pairs and save to file demo.jbl
collect_objective_set(__lowercase , __lowercase , __lowercase , __lowercase , __lowercase , __lowercase , __lowercase , __lowercase)
# clean up, delete model and data we don't need anymore
del model, train_data, objective_set
torch.cuda.empty_cache()
def _snake_case (__lowercase , __lowercase=15 , __lowercase=128 , __lowercase=100 , __lowercase="igf_model.pt" , ):
set_seed(42)
# Load pre-trained model
UpperCamelCase_ = GPTaLMHeadModel.from_pretrained('gpt2')
# Initialize secondary learner to use embedding weights of model
UpperCamelCase_ = SecondaryLearner(__lowercase)
# Train secondary learner
UpperCamelCase_ = train_secondary_learner(
__lowercase , __lowercase , max_epochs=__lowercase , batch_size=__lowercase , eval_freq=100 , igf_model_path=__lowercase , )
del model, secondary_learner_train_data
torch.cuda.empty_cache()
return secondary_learner
def _snake_case (__lowercase , __lowercase , __lowercase , __lowercase=32 , __lowercase=1000 , __lowercase=16 , __lowercase=1.0 , __lowercase=recopy_gpta , __lowercase=None , __lowercase=10 , __lowercase="gpt2_finetuned.pt" , ):
UpperCamelCase_ = torch.device('cuda:0' if torch.cuda.is_available() else 'cpu')
UpperCamelCase_ = RandomSampler(__lowercase)
UpperCamelCase_ = DataLoader(__lowercase , sampler=__lowercase)
UpperCamelCase_ = max_steps // (len(__lowercase)) + 1
UpperCamelCase_ = 0
UpperCamelCase_ = torch.zeros((1, context_len) , dtype=torch.long , device=__lowercase)
UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ = recopy_model(__lowercase , __lowercase , __lowercase)
model.train()
if secondary_learner is not None:
secondary_learner.to(__lowercase)
secondary_learner.eval()
UpperCamelCase_ = []
UpperCamelCase_ = 0
UpperCamelCase_ = []
UpperCamelCase_ = []
# Compute the performance of the transformer model at the beginning
UpperCamelCase_ = compute_perplexity(__lowercase , __lowercase , __lowercase)
test_perps.append(__lowercase)
print('Test perplexity, step' , __lowercase , ':' , __lowercase)
for epoch in range(int(__lowercase)):
for step, example in enumerate(__lowercase):
torch.cuda.empty_cache()
UpperCamelCase_ = random.randint(0 , example.size(2) - context_len - 1)
UpperCamelCase_ = example[0, 0, start : start + context_len]
lm_optimizer.zero_grad()
UpperCamelCase_ = model(__lowercase , labels=__lowercase)
UpperCamelCase_ = True
if secondary_learner is not None:
UpperCamelCase_ = secondary_learner.forward(
torch.tensor(__lowercase , dtype=torch.long , device=__lowercase).unsqueeze(0))[0].item()
observed_qs.append(float(__lowercase))
# Here we implement the simple non-constant threshold for the predicted IG(X) value
# We will decay the selectivity of our secondary learner filter from
# 1 standard deviation above average to 1 below average after 10 batches.
if global_step == 10:
UpperCamelCase_ = -1
if predicted_q < threshold:
UpperCamelCase_ = False
# If we passed the filter, add the context to the batch!
if do_backprop:
contexts.append(np.array(context.cpu()))
UpperCamelCase_ = outputs[0]
lm_loss.backward()
examples += 1
del outputs
# Once the batch is filled with enough contexts, backprop on the batch.
if examples == batch_size:
torch.cuda.empty_cache()
UpperCamelCase_ = 0
# Do LM backprop
torch.nn.utils.clip_grad_norm_(model.parameters() , 3.0)
lm_optimizer.step()
lm_scheduler.step() # Update learning rate schedule
global_step += 1
# Compute the performance of the transformer model at this batch
if global_step % eval_interval == 0:
UpperCamelCase_ = compute_perplexity(__lowercase , __lowercase , __lowercase)
test_perps.append(__lowercase)
print('Test perplexity, step' , __lowercase , ':' , __lowercase)
# Break out of the loop after 60 batches
if max_steps > 0 and global_step > 60:
break
if max_steps > 0 and global_step > 60:
break
# save finetuned transformer model
torch.save(model.state_dict() , __lowercase)
torch.cuda.empty_cache()
# Do some cleaning up so we can reinitialize for the next run of this function
del lm_optimizer
del lm_scheduler
return model
def _snake_case ():
UpperCamelCase_ = argparse.ArgumentParser(description='Fine-tune a transformer model with IGF on a language modeling task')
# Required parameters
parser.add_argument(
'--data_dir' , default=__lowercase , type=__lowercase , required=__lowercase , help='The input data dir. Should contain data files for WikiText.' , )
parser.add_argument(
'--model_name_or_path' , default=__lowercase , type=__lowercase , required=__lowercase , help='Path to pretrained model or model identifier from huggingface.co/models' , )
parser.add_argument(
'--data_file' , type=__lowercase , default=__lowercase , help=(
'A jbl file containing tokenized data which can be split as objective dataset, '
'train_dataset and test_dataset.'
) , )
parser.add_argument(
'--igf_data_file' , type=__lowercase , default=__lowercase , help='A jbl file containing the context and information gain pairs to train secondary learner.' , )
parser.add_argument(
'--output_dir' , default=__lowercase , type=__lowercase , required=__lowercase , help='The output directory where the final fine-tuned model is stored.' , )
parser.add_argument(
'--tokenizer_name' , default=__lowercase , type=__lowercase , help='Pretrained tokenizer name or path if not the same as model_name' , )
parser.add_argument('--seed' , type=__lowercase , default=__lowercase , help='A seed for reproducible training.')
parser.add_argument(
'--context_len' , default=32 , type=__lowercase , help=(
'The maximum total input sequence length after tokenization. Sequences longer '
'than this will be truncated, sequences shorter will be padded.'
) , )
parser.add_argument(
'--size_objective_set' , default=100 , type=__lowercase , help='number of articles that are long enough to be used as our objective set' , )
parser.add_argument(
'--eval_freq' , default=100 , type=__lowercase , help='secondary model evaluation is triggered at eval_freq')
parser.add_argument('--max_steps' , default=1000 , type=__lowercase , help='To calculate training epochs')
parser.add_argument(
'--secondary_learner_batch_size' , default=128 , type=__lowercase , help='batch size of training data for secondary learner' , )
parser.add_argument(
'--batch_size' , default=16 , type=__lowercase , help='batch size of training data of language model(gpt2) ')
parser.add_argument(
'--eval_interval' , default=10 , type=__lowercase , help=(
'decay the selectivity of our secondary learner filter from'
'1 standard deviation above average to 1 below average after 10 batches'
) , )
parser.add_argument(
'--number' , default=100 , type=__lowercase , help='The number of examples split to be used as objective_set/test_data')
parser.add_argument(
'--min_len' , default=1026 , type=__lowercase , help='The minimum length of the article to be used as objective set')
parser.add_argument(
'--secondary_learner_max_epochs' , default=15 , type=__lowercase , help='number of epochs to train secondary learner')
parser.add_argument('--trim' , default=__lowercase , type=__lowercase , help='truncate the example if it exceeds context length')
parser.add_argument(
'--threshold' , default=1.0 , type=__lowercase , help=(
'The threshold value used by secondary learner to filter the train_data and allow only'
' informative data as input to the model'
) , )
parser.add_argument('--finetuned_model_name' , default='gpt2_finetuned.pt' , type=__lowercase , help='finetuned_model_name')
parser.add_argument(
'--recopy_model' , default=__lowercase , type=__lowercase , help='Reset the model to the original pretrained GPT-2 weights after each iteration' , )
# function calls
# Collecting *n* pairs of context and information gain(X, IG(X)) for training the secondary learner
generate_n_pairs(
context_len=32 , max_steps=10 , size_objective_set=100 , min_len=1026 , trim=__lowercase , data_file='data/tokenized_stories_train_wikitext103.jbl' , igf_data_file='igf_context_pairs.jbl' , )
# Load train data for secondary learner
UpperCamelCase_ = joblib.load('data/IGF_values.jbl')
# Train secondary learner
UpperCamelCase_ = training_secondary_learner(
__lowercase , secondary_learner_max_epochs=15 , secondary_learner_batch_size=128 , eval_freq=100 , igf_model_path='igf_model.pt' , )
# load pretrained gpt2 model
UpperCamelCase_ = GPTaLMHeadModel.from_pretrained('gpt2')
set_seed(42)
# Generate train and test data to train and evaluate gpt2 model
UpperCamelCase_ , UpperCamelCase_ = generate_datasets(
context_len=32 , file='data/tokenized_stories_train_wikitext103.jbl' , number=100 , min_len=1026 , trim=__lowercase)
# fine-tuning of the gpt2 model using igf (Information Gain Filtration)
finetune(
__lowercase , __lowercase , __lowercase , context_len=32 , max_steps=1000 , batch_size=16 , threshold=1.0 , recopy_model=__lowercase , secondary_learner=__lowercase , eval_interval=10 , finetuned_model_name='gpt2_finetuned.pt' , )
if __name__ == "__main__":
main()
| 23 | 0 |
from collections import deque
def SCREAMING_SNAKE_CASE ( __lowerCAmelCase ) -> Union[str, Any]:
snake_case__ = len(__lowerCAmelCase )
snake_case__ = deque()
snake_case__ = [False for _ in range(__lowerCAmelCase )]
snake_case__ = [-1 for _ in range(__lowerCAmelCase )]
snake_case__ = index_of[:]
def strong_connect(__lowerCAmelCase , __lowerCAmelCase , __lowerCAmelCase ):
snake_case__ = index # the number when this node is seen
snake_case__ = index # lowest rank node reachable from here
index += 1
stack.append(__lowerCAmelCase )
snake_case__ = True
for w in g[v]:
if index_of[w] == -1:
snake_case__ = strong_connect(__lowerCAmelCase , __lowerCAmelCase , __lowerCAmelCase )
snake_case__ = (
lowlink_of[w] if lowlink_of[w] < lowlink_of[v] else lowlink_of[v]
)
elif on_stack[w]:
snake_case__ = (
lowlink_of[w] if lowlink_of[w] < lowlink_of[v] else lowlink_of[v]
)
if lowlink_of[v] == index_of[v]:
snake_case__ = []
snake_case__ = stack.pop()
snake_case__ = False
component.append(__lowerCAmelCase )
while w != v:
snake_case__ = stack.pop()
snake_case__ = False
component.append(__lowerCAmelCase )
components.append(__lowerCAmelCase )
return index
snake_case__ = []
for v in range(__lowerCAmelCase ):
if index_of[v] == -1:
strong_connect(__lowerCAmelCase , 0 , __lowerCAmelCase )
return components
def SCREAMING_SNAKE_CASE ( __lowerCAmelCase , __lowerCAmelCase ) -> Optional[int]:
snake_case__ = [[] for _ in range(__lowerCAmelCase )]
for u, v in edges:
g[u].append(__lowerCAmelCase )
return g
if __name__ == "__main__":
# Test
lowerCamelCase__ : Tuple = 7
lowerCamelCase__ : Optional[Any] = [0, 0, 1, 2, 3, 3, 4, 4, 6]
lowerCamelCase__ : Optional[int] = [1, 3, 2, 0, 1, 4, 5, 6, 5]
lowerCamelCase__ : int = [(u, v) for u, v in zip(source, target)]
lowerCamelCase__ : List[str] = create_graph(n_vertices, edges)
assert [[5], [6], [4], [3, 2, 1, 0]] == tarjan(g)
| 33 |
from __future__ import annotations
import unittest
from transformers import AutoTokenizer, MBartConfig, is_tf_available
from transformers.testing_utils import require_sentencepiece, require_tf, require_tokenizers, slow
from transformers.utils import cached_property
from ...test_configuration_common import ConfigTester
from ...test_modeling_tf_common import TFModelTesterMixin, ids_tensor
from ...test_pipeline_mixin import PipelineTesterMixin
if is_tf_available():
import tensorflow as tf
from transformers import TFAutoModelForSeqaSeqLM, TFMBartForConditionalGeneration, TFMBartModel
@require_tf
class _a :
"""simple docstring"""
A_ = MBartConfig
A_ = {}
A_ = """gelu"""
def __init__( self , _UpperCAmelCase , _UpperCAmelCase=13 , _UpperCAmelCase=7 , _UpperCAmelCase=True , _UpperCAmelCase=False , _UpperCAmelCase=99 , _UpperCAmelCase=32 , _UpperCAmelCase=2 , _UpperCAmelCase=4 , _UpperCAmelCase=37 , _UpperCAmelCase=0.1 , _UpperCAmelCase=0.1 , _UpperCAmelCase=20 , _UpperCAmelCase=2 , _UpperCAmelCase=1 , _UpperCAmelCase=0 , ) -> Union[str, Any]:
UpperCamelCase_ = parent
UpperCamelCase_ = batch_size
UpperCamelCase_ = seq_length
UpperCamelCase_ = is_training
UpperCamelCase_ = use_labels
UpperCamelCase_ = vocab_size
UpperCamelCase_ = hidden_size
UpperCamelCase_ = num_hidden_layers
UpperCamelCase_ = num_attention_heads
UpperCamelCase_ = intermediate_size
UpperCamelCase_ = hidden_dropout_prob
UpperCamelCase_ = attention_probs_dropout_prob
UpperCamelCase_ = max_position_embeddings
UpperCamelCase_ = eos_token_id
UpperCamelCase_ = pad_token_id
UpperCamelCase_ = bos_token_id
def _UpperCAmelCase ( self ) -> List[Any]:
UpperCamelCase_ = ids_tensor([self.batch_size, self.seq_length - 1] , self.vocab_size )
UpperCamelCase_ = tf.expand_dims(tf.constant([self.eos_token_id] * self.batch_size ) , 1 )
UpperCamelCase_ = tf.concat([input_ids, eos_tensor] , axis=1 )
UpperCamelCase_ = ids_tensor([self.batch_size, self.seq_length] , self.vocab_size )
UpperCamelCase_ = self.config_cls(
vocab_size=self.vocab_size , d_model=self.hidden_size , encoder_layers=self.num_hidden_layers , decoder_layers=self.num_hidden_layers , encoder_attention_heads=self.num_attention_heads , decoder_attention_heads=self.num_attention_heads , encoder_ffn_dim=self.intermediate_size , decoder_ffn_dim=self.intermediate_size , dropout=self.hidden_dropout_prob , attention_dropout=self.attention_probs_dropout_prob , max_position_embeddings=self.max_position_embeddings , eos_token_ids=[2] , bos_token_id=self.bos_token_id , pad_token_id=self.pad_token_id , decoder_start_token_id=self.pad_token_id , **self.config_updates , )
UpperCamelCase_ = prepare_mbart_inputs_dict(_UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase )
return config, inputs_dict
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> List[str]:
UpperCamelCase_ = TFMBartModel(config=_UpperCAmelCase ).get_decoder()
UpperCamelCase_ = inputs_dict['input_ids']
UpperCamelCase_ = input_ids[:1, :]
UpperCamelCase_ = inputs_dict['attention_mask'][:1, :]
UpperCamelCase_ = inputs_dict['head_mask']
UpperCamelCase_ = 1
# first forward pass
UpperCamelCase_ = model(_UpperCAmelCase , attention_mask=_UpperCAmelCase , head_mask=_UpperCAmelCase , use_cache=_UpperCAmelCase )
UpperCamelCase_ , UpperCamelCase_ = outputs.to_tuple()
UpperCamelCase_ = past_key_values[1]
def _snake_case (__lowercase , __lowercase , __lowercase , __lowercase=None , __lowercase=None , __lowercase=None , __lowercase=None , __lowercase=None , ):
if attention_mask is None:
UpperCamelCase_ = tf.cast(tf.math.not_equal(__lowercase , config.pad_token_id) , tf.inta)
if decoder_attention_mask is None:
UpperCamelCase_ = tf.concat(
[
tf.ones(decoder_input_ids[:, :1].shape , dtype=tf.inta),
tf.cast(tf.math.not_equal(decoder_input_ids[:, 1:] , config.pad_token_id) , tf.inta),
] , axis=-1 , )
if head_mask is None:
UpperCamelCase_ = tf.ones((config.encoder_layers, config.encoder_attention_heads))
if decoder_head_mask is None:
UpperCamelCase_ = tf.ones((config.decoder_layers, config.decoder_attention_heads))
if cross_attn_head_mask is None:
UpperCamelCase_ = tf.ones((config.decoder_layers, config.decoder_attention_heads))
return {
"input_ids": input_ids,
"decoder_input_ids": decoder_input_ids,
"attention_mask": attention_mask,
"decoder_attention_mask": decoder_attention_mask,
"head_mask": head_mask,
"decoder_head_mask": decoder_head_mask,
"cross_attn_head_mask": cross_attn_head_mask,
}
@require_tf
class _a ( UpperCAmelCase__ , UpperCAmelCase__ , unittest.TestCase ):
"""simple docstring"""
A_ = (TFMBartForConditionalGeneration, TFMBartModel) if is_tf_available() else ()
A_ = (TFMBartForConditionalGeneration,) if is_tf_available() else ()
A_ = (
{
"""conversational""": TFMBartForConditionalGeneration,
"""feature-extraction""": TFMBartModel,
"""summarization""": TFMBartForConditionalGeneration,
"""text2text-generation""": TFMBartForConditionalGeneration,
"""translation""": TFMBartForConditionalGeneration,
}
if is_tf_available()
else {}
)
A_ = True
A_ = False
A_ = False
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase ) -> Tuple:
if pipeline_test_casse_name != "FeatureExtractionPipelineTests":
# Exception encountered when calling layer '...'
return True
return False
def _UpperCAmelCase ( self ) -> Optional[int]:
UpperCamelCase_ = TFMBartModelTester(self )
UpperCamelCase_ = ConfigTester(self , config_class=_UpperCAmelCase )
def _UpperCAmelCase ( self ) -> Optional[int]:
self.config_tester.run_common_tests()
def _UpperCAmelCase ( self ) -> List[Any]:
UpperCamelCase_ = self.model_tester.prepare_config_and_inputs_for_common()
self.model_tester.check_decoder_model_past_large_inputs(*_UpperCAmelCase )
@require_sentencepiece
@require_tokenizers
@require_tf
class _a ( unittest.TestCase ):
"""simple docstring"""
A_ = [
""" UN Chief Says There Is No Military Solution in Syria""",
]
A_ = [
"""Şeful ONU declară că nu există o soluţie militară în Siria""",
]
A_ = """facebook/mbart-large-en-ro"""
@cached_property
def _UpperCAmelCase ( self ) -> Any:
return AutoTokenizer.from_pretrained(self.model_name )
@cached_property
def _UpperCAmelCase ( self ) -> List[str]:
UpperCamelCase_ = TFAutoModelForSeqaSeqLM.from_pretrained(self.model_name )
return model
def _UpperCAmelCase ( self , **_UpperCAmelCase ) -> int:
UpperCamelCase_ = self.translate_src_text(**_UpperCAmelCase )
self.assertListEqual(self.expected_text , _UpperCAmelCase )
def _UpperCAmelCase ( self , **_UpperCAmelCase ) -> List[str]:
UpperCamelCase_ = self.tokenizer(self.src_text , **_UpperCAmelCase , return_tensors='tf' )
UpperCamelCase_ = self.model.generate(
model_inputs.input_ids , attention_mask=model_inputs.attention_mask , num_beams=2 )
UpperCamelCase_ = self.tokenizer.batch_decode(_UpperCAmelCase , skip_special_tokens=_UpperCAmelCase )
return generated_words
@slow
def _UpperCAmelCase ( self ) -> List[Any]:
self._assert_generated_batch_equal_expected()
| 23 | 0 |
"""simple docstring"""
import re
import string
import numpy as np
import datasets
SCREAMING_SNAKE_CASE_ = '\nReturns the rate at which the input predicted strings exactly match their references, ignoring any strings input as part of the regexes_to_ignore list.\n'
SCREAMING_SNAKE_CASE_ = '\nArgs:\n predictions: List of predicted texts.\n references: List of reference texts.\n regexes_to_ignore: List, defaults to None. Regex expressions of characters to\n ignore when calculating the exact matches. Note: these regexes are removed\n from the input data before the changes based on the options below (e.g. ignore_case,\n ignore_punctuation, ignore_numbers) are applied.\n ignore_case: Boolean, defaults to False. If true, turns everything\n to lowercase so that capitalization differences are ignored.\n ignore_punctuation: Boolean, defaults to False. If true, removes all punctuation before\n comparing predictions and references.\n ignore_numbers: Boolean, defaults to False. If true, removes all punctuation before\n comparing predictions and references.\nReturns:\n exact_match: Dictionary containing exact_match rate. Possible values are between 0.0 and 100.0, inclusive.\nExamples:\n >>> exact_match = datasets.load_metric("exact_match")\n >>> refs = ["the cat", "theater", "YELLING", "agent007"]\n >>> preds = ["cat?", "theater", "yelling", "agent"]\n >>> results = exact_match.compute(references=refs, predictions=preds)\n >>> print(round(results["exact_match"], 1))\n 25.0\n\n >>> exact_match = datasets.load_metric("exact_match")\n >>> refs = ["the cat", "theater", "YELLING", "agent007"]\n >>> preds = ["cat?", "theater", "yelling", "agent"]\n >>> results = exact_match.compute(references=refs, predictions=preds, regexes_to_ignore=["the ", "yell"], ignore_case=True, ignore_punctuation=True)\n >>> print(round(results["exact_match"], 1))\n 50.0\n\n\n >>> exact_match = datasets.load_metric("exact_match")\n >>> refs = ["the cat", "theater", "YELLING", "agent007"]\n >>> preds = ["cat?", "theater", "yelling", "agent"]\n >>> results = exact_match.compute(references=refs, predictions=preds, regexes_to_ignore=["the ", "yell", "YELL"], ignore_case=True, ignore_punctuation=True)\n >>> print(round(results["exact_match"], 1))\n 75.0\n\n >>> exact_match = datasets.load_metric("exact_match")\n >>> refs = ["the cat", "theater", "YELLING", "agent007"]\n >>> preds = ["cat?", "theater", "yelling", "agent"]\n >>> results = exact_match.compute(references=refs, predictions=preds, regexes_to_ignore=["the ", "yell", "YELL"], ignore_case=True, ignore_punctuation=True, ignore_numbers=True)\n >>> print(round(results["exact_match"], 1))\n 100.0\n\n >>> exact_match = datasets.load_metric("exact_match")\n >>> refs = ["The cat sat on the mat.", "Theaters are great.", "It\'s like comparing oranges and apples."]\n >>> preds = ["The cat sat on the mat?", "Theaters are great.", "It\'s like comparing apples and oranges."]\n >>> results = exact_match.compute(references=refs, predictions=preds)\n >>> print(round(results["exact_match"], 1))\n 33.3\n\n'
SCREAMING_SNAKE_CASE_ = '\n'
@datasets.utils.file_utils.add_start_docstrings(_DESCRIPTION , _KWARGS_DESCRIPTION )
class snake_case_ ( datasets.Metric ):
"""simple docstring"""
def UpperCAmelCase__ ( self) -> str:
return datasets.MetricInfo(
description=_DESCRIPTION , citation=_CITATION , inputs_description=_KWARGS_DESCRIPTION , features=datasets.Features(
{
'''predictions''': datasets.Value('''string''' , id='''sequence'''),
'''references''': datasets.Value('''string''' , id='''sequence'''),
}) , reference_urls=[] , )
def UpperCAmelCase__ ( self , lowerCamelCase_ , lowerCamelCase_ , lowerCamelCase_=None , lowerCamelCase_=False , lowerCamelCase_=False , lowerCamelCase_=False , ) -> Any:
if regexes_to_ignore is not None:
for s in regexes_to_ignore:
UpperCamelCase = np.array([re.sub(lowerCamelCase_ , '''''' , lowerCamelCase_) for x in predictions])
UpperCamelCase = np.array([re.sub(lowerCamelCase_ , '''''' , lowerCamelCase_) for x in references])
else:
UpperCamelCase = np.asarray(lowerCamelCase_)
UpperCamelCase = np.asarray(lowerCamelCase_)
if ignore_case:
UpperCamelCase = np.char.lower(lowerCamelCase_)
UpperCamelCase = np.char.lower(lowerCamelCase_)
if ignore_punctuation:
UpperCamelCase = string.punctuation.maketrans('''''' , '''''' , string.punctuation)
UpperCamelCase = np.char.translate(lowerCamelCase_ , table=lowerCamelCase_)
UpperCamelCase = np.char.translate(lowerCamelCase_ , table=lowerCamelCase_)
if ignore_numbers:
UpperCamelCase = string.digits.maketrans('''''' , '''''' , string.digits)
UpperCamelCase = np.char.translate(lowerCamelCase_ , table=lowerCamelCase_)
UpperCamelCase = np.char.translate(lowerCamelCase_ , table=lowerCamelCase_)
UpperCamelCase = predictions == references
return {"exact_match": np.mean(lowerCamelCase_) * 1_0_0} | 34 |
def _snake_case (__lowercase):
UpperCamelCase_ = 1
for i in range(1 , num + 1):
fact *= i
return fact
def _snake_case (__lowercase):
UpperCamelCase_ = 0
while number > 0:
UpperCamelCase_ = number % 10
sum_of_digits += last_digit
UpperCamelCase_ = number // 10 # Removing the last_digit from the given number
return sum_of_digits
def _snake_case (__lowercase = 100):
UpperCamelCase_ = factorial(__lowercase)
UpperCamelCase_ = split_and_add(__lowercase)
return result
if __name__ == "__main__":
print(solution(int(input("""Enter the Number: """).strip())))
| 23 | 0 |
from ...utils import (
OptionalDependencyNotAvailable,
is_torch_available,
is_transformers_available,
is_transformers_version,
)
try:
if not (is_transformers_available() and is_torch_available()):
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
from ...utils.dummy_torch_and_transformers_objects import ShapEPipeline
else:
from .camera import create_pan_cameras
from .pipeline_shap_e import ShapEPipeline
from .pipeline_shap_e_img2img import ShapEImgaImgPipeline
from .renderer import (
BoundingBoxVolume,
ImportanceRaySampler,
MLPNeRFModelOutput,
MLPNeRSTFModel,
ShapEParamsProjModel,
ShapERenderer,
StratifiedRaySampler,
VoidNeRFModel,
)
| 35 |
from typing import Dict, List, Optional, Union
import numpy as np
from ...image_processing_utils import BaseImageProcessor, BatchFeature, get_size_dict
from ...image_transforms import (
center_crop,
get_resize_output_image_size,
normalize,
rescale,
resize,
to_channel_dimension_format,
)
from ...image_utils import (
IMAGENET_STANDARD_MEAN,
IMAGENET_STANDARD_STD,
ChannelDimension,
ImageInput,
PILImageResampling,
is_valid_image,
to_numpy_array,
valid_images,
)
from ...utils import TensorType, is_vision_available, logging
if is_vision_available():
import PIL
snake_case__ : str = logging.get_logger(__name__)
def _snake_case (__lowercase):
if isinstance(__lowercase , (list, tuple)) and isinstance(videos[0] , (list, tuple)) and is_valid_image(videos[0][0]):
return videos
elif isinstance(__lowercase , (list, tuple)) and is_valid_image(videos[0]):
return [videos]
elif is_valid_image(__lowercase):
return [[videos]]
raise ValueError(f"""Could not make batched video from {videos}""")
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
A_ = ["""pixel_values"""]
def __init__( self , _UpperCAmelCase = True , _UpperCAmelCase = None , _UpperCAmelCase = PILImageResampling.BILINEAR , _UpperCAmelCase = True , _UpperCAmelCase = None , _UpperCAmelCase = True , _UpperCAmelCase = 1 / 255 , _UpperCAmelCase = True , _UpperCAmelCase = None , _UpperCAmelCase = None , **_UpperCAmelCase , ) -> None:
super().__init__(**_UpperCAmelCase )
UpperCamelCase_ = size if size is not None else {'shortest_edge': 224}
UpperCamelCase_ = get_size_dict(_UpperCAmelCase , default_to_square=_UpperCAmelCase )
UpperCamelCase_ = crop_size if crop_size is not None else {'height': 224, 'width': 224}
UpperCamelCase_ = get_size_dict(_UpperCAmelCase , param_name='crop_size' )
UpperCamelCase_ = do_resize
UpperCamelCase_ = size
UpperCamelCase_ = do_center_crop
UpperCamelCase_ = crop_size
UpperCamelCase_ = resample
UpperCamelCase_ = do_rescale
UpperCamelCase_ = rescale_factor
UpperCamelCase_ = do_normalize
UpperCamelCase_ = image_mean if image_mean is not None else IMAGENET_STANDARD_MEAN
UpperCamelCase_ = image_std if image_std is not None else IMAGENET_STANDARD_STD
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase = PILImageResampling.BILINEAR , _UpperCAmelCase = None , **_UpperCAmelCase , ) -> np.ndarray:
UpperCamelCase_ = get_size_dict(_UpperCAmelCase , default_to_square=_UpperCAmelCase )
if "shortest_edge" in size:
UpperCamelCase_ = get_resize_output_image_size(_UpperCAmelCase , size['shortest_edge'] , default_to_square=_UpperCAmelCase )
elif "height" in size and "width" in size:
UpperCamelCase_ = (size['height'], size['width'])
else:
raise ValueError(f"""Size must have 'height' and 'width' or 'shortest_edge' as keys. Got {size.keys()}""" )
return resize(_UpperCAmelCase , size=_UpperCAmelCase , resample=_UpperCAmelCase , data_format=_UpperCAmelCase , **_UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase = None , **_UpperCAmelCase , ) -> np.ndarray:
UpperCamelCase_ = get_size_dict(_UpperCAmelCase )
if "height" not in size or "width" not in size:
raise ValueError(f"""Size must have 'height' and 'width' as keys. Got {size.keys()}""" )
return center_crop(_UpperCAmelCase , size=(size['height'], size['width']) , data_format=_UpperCAmelCase , **_UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase = None , **_UpperCAmelCase , ) -> int:
return rescale(_UpperCAmelCase , scale=_UpperCAmelCase , data_format=_UpperCAmelCase , **_UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase = None , **_UpperCAmelCase , ) -> np.ndarray:
return normalize(_UpperCAmelCase , mean=_UpperCAmelCase , std=_UpperCAmelCase , data_format=_UpperCAmelCase , **_UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = ChannelDimension.FIRST , ) -> np.ndarray:
if do_resize and size is None or resample is None:
raise ValueError('Size and resample must be specified if do_resize is True.' )
if do_center_crop and crop_size is None:
raise ValueError('Crop size must be specified if do_center_crop is True.' )
if do_rescale and rescale_factor is None:
raise ValueError('Rescale factor must be specified if do_rescale is True.' )
if do_normalize and (image_mean is None or image_std is None):
raise ValueError('Image mean and std must be specified if do_normalize is True.' )
# All transformations expect numpy arrays.
UpperCamelCase_ = to_numpy_array(_UpperCAmelCase )
if do_resize:
UpperCamelCase_ = self.resize(image=_UpperCAmelCase , size=_UpperCAmelCase , resample=_UpperCAmelCase )
if do_center_crop:
UpperCamelCase_ = self.center_crop(_UpperCAmelCase , size=_UpperCAmelCase )
if do_rescale:
UpperCamelCase_ = self.rescale(image=_UpperCAmelCase , scale=_UpperCAmelCase )
if do_normalize:
UpperCamelCase_ = self.normalize(image=_UpperCAmelCase , mean=_UpperCAmelCase , std=_UpperCAmelCase )
UpperCamelCase_ = to_channel_dimension_format(_UpperCAmelCase , _UpperCAmelCase )
return image
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = ChannelDimension.FIRST , **_UpperCAmelCase , ) -> PIL.Image.Image:
UpperCamelCase_ = do_resize if do_resize is not None else self.do_resize
UpperCamelCase_ = resample if resample is not None else self.resample
UpperCamelCase_ = do_center_crop if do_center_crop is not None else self.do_center_crop
UpperCamelCase_ = do_rescale if do_rescale is not None else self.do_rescale
UpperCamelCase_ = rescale_factor if rescale_factor is not None else self.rescale_factor
UpperCamelCase_ = do_normalize if do_normalize is not None else self.do_normalize
UpperCamelCase_ = image_mean if image_mean is not None else self.image_mean
UpperCamelCase_ = image_std if image_std is not None else self.image_std
UpperCamelCase_ = size if size is not None else self.size
UpperCamelCase_ = get_size_dict(_UpperCAmelCase , default_to_square=_UpperCAmelCase )
UpperCamelCase_ = crop_size if crop_size is not None else self.crop_size
UpperCamelCase_ = get_size_dict(_UpperCAmelCase , param_name='crop_size' )
if not valid_images(_UpperCAmelCase ):
raise ValueError(
'Invalid image type. Must be of type PIL.Image.Image, numpy.ndarray, '
'torch.Tensor, tf.Tensor or jax.ndarray.' )
UpperCamelCase_ = make_batched(_UpperCAmelCase )
UpperCamelCase_ = [
[
self._preprocess_image(
image=_UpperCAmelCase , do_resize=_UpperCAmelCase , size=_UpperCAmelCase , resample=_UpperCAmelCase , do_center_crop=_UpperCAmelCase , crop_size=_UpperCAmelCase , do_rescale=_UpperCAmelCase , rescale_factor=_UpperCAmelCase , do_normalize=_UpperCAmelCase , image_mean=_UpperCAmelCase , image_std=_UpperCAmelCase , data_format=_UpperCAmelCase , )
for img in video
]
for video in videos
]
UpperCamelCase_ = {'pixel_values': videos}
return BatchFeature(data=_UpperCAmelCase , tensor_type=_UpperCAmelCase )
| 23 | 0 |
import numpy as np
def lowercase ( __A : np.array ) -> np.array:
'''simple docstring'''
return (2 / (1 + np.exp(-2 * vector ))) - 1
if __name__ == "__main__":
import doctest
doctest.testmod()
| 36 |
# DISCLAIMER: This file is strongly influenced by https://github.com/yang-song/score_sde_pytorch
import math
from dataclasses import dataclass
from typing import Optional, Tuple, Union
import torch
from ..configuration_utils import ConfigMixin, register_to_config
from ..utils import BaseOutput, randn_tensor
from .scheduling_utils import SchedulerMixin, SchedulerOutput
@dataclass
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
A_ = 42
A_ = 42
class _a ( UpperCAmelCase__ , UpperCAmelCase__ ):
"""simple docstring"""
A_ = 1
@register_to_config
def __init__( self , _UpperCAmelCase = 2000 , _UpperCAmelCase = 0.1_5 , _UpperCAmelCase = 0.0_1 , _UpperCAmelCase = 1_3_4_8.0 , _UpperCAmelCase = 1e-5 , _UpperCAmelCase = 1 , ) -> Tuple:
# standard deviation of the initial noise distribution
UpperCamelCase_ = sigma_max
# setable values
UpperCamelCase_ = None
self.set_sigmas(_UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase = None ) -> torch.FloatTensor:
return sample
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase = None , _UpperCAmelCase = None ) -> str:
UpperCamelCase_ = sampling_eps if sampling_eps is not None else self.config.sampling_eps
UpperCamelCase_ = torch.linspace(1 , _UpperCAmelCase , _UpperCAmelCase , device=_UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None ) -> Any:
UpperCamelCase_ = sigma_min if sigma_min is not None else self.config.sigma_min
UpperCamelCase_ = sigma_max if sigma_max is not None else self.config.sigma_max
UpperCamelCase_ = sampling_eps if sampling_eps is not None else self.config.sampling_eps
if self.timesteps is None:
self.set_timesteps(_UpperCAmelCase , _UpperCAmelCase )
UpperCamelCase_ = sigma_min * (sigma_max / sigma_min) ** (self.timesteps / sampling_eps)
UpperCamelCase_ = torch.exp(torch.linspace(math.log(_UpperCAmelCase ) , math.log(_UpperCAmelCase ) , _UpperCAmelCase ) )
UpperCamelCase_ = torch.tensor([sigma_min * (sigma_max / sigma_min) ** t for t in self.timesteps] )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> List[Any]:
return torch.where(
timesteps == 0 , torch.zeros_like(t.to(timesteps.device ) ) , self.discrete_sigmas[timesteps - 1].to(timesteps.device ) , )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase = None , _UpperCAmelCase = True , ) -> Union[SdeVeOutput, Tuple]:
if self.timesteps is None:
raise ValueError(
'`self.timesteps` is not set, you need to run \'set_timesteps\' after creating the scheduler' )
UpperCamelCase_ = timestep * torch.ones(
sample.shape[0] , device=sample.device ) # torch.repeat_interleave(timestep, sample.shape[0])
UpperCamelCase_ = (timestep * (len(self.timesteps ) - 1)).long()
# mps requires indices to be in the same device, so we use cpu as is the default with cuda
UpperCamelCase_ = timesteps.to(self.discrete_sigmas.device )
UpperCamelCase_ = self.discrete_sigmas[timesteps].to(sample.device )
UpperCamelCase_ = self.get_adjacent_sigma(_UpperCAmelCase , _UpperCAmelCase ).to(sample.device )
UpperCamelCase_ = torch.zeros_like(_UpperCAmelCase )
UpperCamelCase_ = (sigma**2 - adjacent_sigma**2) ** 0.5
# equation 6 in the paper: the model_output modeled by the network is grad_x log pt(x)
# also equation 47 shows the analog from SDE models to ancestral sampling methods
UpperCamelCase_ = diffusion.flatten()
while len(diffusion.shape ) < len(sample.shape ):
UpperCamelCase_ = diffusion.unsqueeze(-1 )
UpperCamelCase_ = drift - diffusion**2 * model_output
# equation 6: sample noise for the diffusion term of
UpperCamelCase_ = randn_tensor(
sample.shape , layout=sample.layout , generator=_UpperCAmelCase , device=sample.device , dtype=sample.dtype )
UpperCamelCase_ = sample - drift # subtract because `dt` is a small negative timestep
# TODO is the variable diffusion the correct scaling term for the noise?
UpperCamelCase_ = prev_sample_mean + diffusion * noise # add impact of diffusion field g
if not return_dict:
return (prev_sample, prev_sample_mean)
return SdeVeOutput(prev_sample=_UpperCAmelCase , prev_sample_mean=_UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase = None , _UpperCAmelCase = True , ) -> Union[SchedulerOutput, Tuple]:
if self.timesteps is None:
raise ValueError(
'`self.timesteps` is not set, you need to run \'set_timesteps\' after creating the scheduler' )
# For small batch sizes, the paper "suggest replacing norm(z) with sqrt(d), where d is the dim. of z"
# sample noise for correction
UpperCamelCase_ = randn_tensor(sample.shape , layout=sample.layout , generator=_UpperCAmelCase ).to(sample.device )
# compute step size from the model_output, the noise, and the snr
UpperCamelCase_ = torch.norm(model_output.reshape(model_output.shape[0] , -1 ) , dim=-1 ).mean()
UpperCamelCase_ = torch.norm(noise.reshape(noise.shape[0] , -1 ) , dim=-1 ).mean()
UpperCamelCase_ = (self.config.snr * noise_norm / grad_norm) ** 2 * 2
UpperCamelCase_ = step_size * torch.ones(sample.shape[0] ).to(sample.device )
# self.repeat_scalar(step_size, sample.shape[0])
# compute corrected sample: model_output term and noise term
UpperCamelCase_ = step_size.flatten()
while len(step_size.shape ) < len(sample.shape ):
UpperCamelCase_ = step_size.unsqueeze(-1 )
UpperCamelCase_ = sample + step_size * model_output
UpperCamelCase_ = prev_sample_mean + ((step_size * 2) ** 0.5) * noise
if not return_dict:
return (prev_sample,)
return SchedulerOutput(prev_sample=_UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , ) -> torch.FloatTensor:
# Make sure sigmas and timesteps have the same device and dtype as original_samples
UpperCamelCase_ = timesteps.to(original_samples.device )
UpperCamelCase_ = self.discrete_sigmas.to(original_samples.device )[timesteps]
UpperCamelCase_ = (
noise * sigmas[:, None, None, None]
if noise is not None
else torch.randn_like(_UpperCAmelCase ) * sigmas[:, None, None, None]
)
UpperCamelCase_ = noise + original_samples
return noisy_samples
def __len__( self ) -> Optional[int]:
return self.config.num_train_timesteps
| 23 | 0 |
from collections import defaultdict
from math import gcd
def UpperCamelCase_ ( __a = 1_500_000 ) -> int:
a__ : defaultdict = defaultdict(__a )
a__ : Optional[int] = 2
while 2 * euclid_m * (euclid_m + 1) <= limit:
for euclid_n in range((euclid_m % 2) + 1 , __a , 2 ):
if gcd(__a , __a ) > 1:
continue
a__ : Any = 2 * euclid_m * (euclid_m + euclid_n)
for perimeter in range(__a , limit + 1 , __a ):
frequencies[perimeter] += 1
euclid_m += 1
return sum(1 for frequency in frequencies.values() if frequency == 1 )
if __name__ == "__main__":
print(f"""{solution() = }""")
| 37 |
from typing import TYPE_CHECKING
from ...utils import OptionalDependencyNotAvailable, _LazyModule, is_torch_available
snake_case__ : Optional[int] = {
"""configuration_pegasus_x""": ["""PEGASUS_X_PRETRAINED_CONFIG_ARCHIVE_MAP""", """PegasusXConfig"""],
}
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
snake_case__ : Dict = [
"""PEGASUS_X_PRETRAINED_MODEL_ARCHIVE_LIST""",
"""PegasusXForConditionalGeneration""",
"""PegasusXModel""",
"""PegasusXPreTrainedModel""",
]
if TYPE_CHECKING:
from .configuration_pegasus_x import PEGASUS_X_PRETRAINED_CONFIG_ARCHIVE_MAP, PegasusXConfig
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_pegasus_x import (
PEGASUS_X_PRETRAINED_MODEL_ARCHIVE_LIST,
PegasusXForConditionalGeneration,
PegasusXModel,
PegasusXPreTrainedModel,
)
else:
import sys
snake_case__ : Optional[Any] = _LazyModule(__name__, globals()["""__file__"""], _import_structure, module_spec=__spec__)
| 23 | 0 |
'''simple docstring'''
from collections import OrderedDict
from typing import Mapping
from ...configuration_utils import PretrainedConfig
from ...onnx import OnnxConfig
from ...utils import logging
A_ : Dict = logging.get_logger(__name__)
A_ : Optional[Any] = {
"xlm-mlm-en-2048": "https://huggingface.co/xlm-mlm-en-2048/resolve/main/config.json",
"xlm-mlm-ende-1024": "https://huggingface.co/xlm-mlm-ende-1024/resolve/main/config.json",
"xlm-mlm-enfr-1024": "https://huggingface.co/xlm-mlm-enfr-1024/resolve/main/config.json",
"xlm-mlm-enro-1024": "https://huggingface.co/xlm-mlm-enro-1024/resolve/main/config.json",
"xlm-mlm-tlm-xnli15-1024": "https://huggingface.co/xlm-mlm-tlm-xnli15-1024/resolve/main/config.json",
"xlm-mlm-xnli15-1024": "https://huggingface.co/xlm-mlm-xnli15-1024/resolve/main/config.json",
"xlm-clm-enfr-1024": "https://huggingface.co/xlm-clm-enfr-1024/resolve/main/config.json",
"xlm-clm-ende-1024": "https://huggingface.co/xlm-clm-ende-1024/resolve/main/config.json",
"xlm-mlm-17-1280": "https://huggingface.co/xlm-mlm-17-1280/resolve/main/config.json",
"xlm-mlm-100-1280": "https://huggingface.co/xlm-mlm-100-1280/resolve/main/config.json",
}
class __snake_case ( __SCREAMING_SNAKE_CASE ):
'''simple docstring'''
lowerCamelCase__ = '''xlm'''
lowerCamelCase__ = {
'''hidden_size''': '''emb_dim''',
'''num_attention_heads''': '''n_heads''',
'''num_hidden_layers''': '''n_layers''',
'''n_words''': '''vocab_size''', # For backward compatibility
}
def __init__( self , __SCREAMING_SNAKE_CASE=3_0_1_4_5 , __SCREAMING_SNAKE_CASE=2_0_4_8 , __SCREAMING_SNAKE_CASE=1_2 , __SCREAMING_SNAKE_CASE=1_6 , __SCREAMING_SNAKE_CASE=0.1 , __SCREAMING_SNAKE_CASE=0.1 , __SCREAMING_SNAKE_CASE=True , __SCREAMING_SNAKE_CASE=False , __SCREAMING_SNAKE_CASE=False , __SCREAMING_SNAKE_CASE=False , __SCREAMING_SNAKE_CASE=1 , __SCREAMING_SNAKE_CASE=True , __SCREAMING_SNAKE_CASE=5_1_2 , __SCREAMING_SNAKE_CASE=2_0_4_8**-0.5 , __SCREAMING_SNAKE_CASE=1e-1_2 , __SCREAMING_SNAKE_CASE=0.02 , __SCREAMING_SNAKE_CASE=0 , __SCREAMING_SNAKE_CASE=1 , __SCREAMING_SNAKE_CASE=2 , __SCREAMING_SNAKE_CASE=3 , __SCREAMING_SNAKE_CASE=5 , __SCREAMING_SNAKE_CASE=True , __SCREAMING_SNAKE_CASE="first" , __SCREAMING_SNAKE_CASE=True , __SCREAMING_SNAKE_CASE=None , __SCREAMING_SNAKE_CASE=True , __SCREAMING_SNAKE_CASE=0.1 , __SCREAMING_SNAKE_CASE=5 , __SCREAMING_SNAKE_CASE=5 , __SCREAMING_SNAKE_CASE=0 , __SCREAMING_SNAKE_CASE=0 , __SCREAMING_SNAKE_CASE=2 , __SCREAMING_SNAKE_CASE=0 , **__SCREAMING_SNAKE_CASE , ):
snake_case__ : Tuple = vocab_size
snake_case__ : Any = emb_dim
snake_case__ : str = n_layers
snake_case__ : Dict = n_heads
snake_case__ : Union[str, Any] = dropout
snake_case__ : Union[str, Any] = attention_dropout
snake_case__ : str = gelu_activation
snake_case__ : List[str] = sinusoidal_embeddings
snake_case__ : Optional[Any] = causal
snake_case__ : int = asm
snake_case__ : List[str] = n_langs
snake_case__ : Any = use_lang_emb
snake_case__ : List[str] = layer_norm_eps
snake_case__ : Optional[Any] = bos_index
snake_case__ : Optional[int] = eos_index
snake_case__ : str = pad_index
snake_case__ : Optional[Any] = unk_index
snake_case__ : Tuple = mask_index
snake_case__ : Tuple = is_encoder
snake_case__ : Any = max_position_embeddings
snake_case__ : Tuple = embed_init_std
snake_case__ : int = init_std
snake_case__ : List[str] = summary_type
snake_case__ : List[str] = summary_use_proj
snake_case__ : int = summary_activation
snake_case__ : Union[str, Any] = summary_proj_to_labels
snake_case__ : Tuple = summary_first_dropout
snake_case__ : List[Any] = start_n_top
snake_case__ : Tuple = end_n_top
snake_case__ : Union[str, Any] = mask_token_id
snake_case__ : Dict = lang_id
if "n_words" in kwargs:
snake_case__ : str = kwargs["""n_words"""]
super().__init__(pad_token_id=__SCREAMING_SNAKE_CASE , bos_token_id=__SCREAMING_SNAKE_CASE , **__SCREAMING_SNAKE_CASE )
class __snake_case ( __SCREAMING_SNAKE_CASE ):
'''simple docstring'''
@property
def __UpperCamelCase ( self ):
if self.task == "multiple-choice":
snake_case__ : Any = {0: """batch""", 1: """choice""", 2: """sequence"""}
else:
snake_case__ : int = {0: """batch""", 1: """sequence"""}
return OrderedDict(
[
("""input_ids""", dynamic_axis),
("""attention_mask""", dynamic_axis),
("""token_type_ids""", dynamic_axis),
] )
| 38 |
import datasets
from .evaluate import evaluate
snake_case__ : int = """\
@article{hendrycks2021cuad,
title={CUAD: An Expert-Annotated NLP Dataset for Legal Contract Review},
author={Dan Hendrycks and Collin Burns and Anya Chen and Spencer Ball},
journal={arXiv preprint arXiv:2103.06268},
year={2021}
}
"""
snake_case__ : Union[str, Any] = """
This metric wrap the official scoring script for version 1 of the Contract
Understanding Atticus Dataset (CUAD).
Contract Understanding Atticus Dataset (CUAD) v1 is a corpus of more than 13,000 labels in 510
commercial legal contracts that have been manually labeled to identify 41 categories of important
clauses that lawyers look for when reviewing contracts in connection with corporate transactions.
"""
snake_case__ : Any = """
Computes CUAD scores (EM, F1, AUPR, Precision@80%Recall, and Precision@90%Recall).
Args:
predictions: List of question-answers dictionaries with the following key-values:
- 'id': id of the question-answer pair as given in the references (see below)
- 'prediction_text': list of possible texts for the answer, as a list of strings
depending on a threshold on the confidence probability of each prediction.
references: List of question-answers dictionaries with the following key-values:
- 'id': id of the question-answer pair (see above),
- 'answers': a Dict in the CUAD dataset format
{
'text': list of possible texts for the answer, as a list of strings
'answer_start': list of start positions for the answer, as a list of ints
}
Note that answer_start values are not taken into account to compute the metric.
Returns:
'exact_match': Exact match (the normalized answer exactly match the gold answer)
'f1': The F-score of predicted tokens versus the gold answer
'aupr': Area Under the Precision-Recall curve
'prec_at_80_recall': Precision at 80% recall
'prec_at_90_recall': Precision at 90% recall
Examples:
>>> predictions = [{'prediction_text': ['The seller:', 'The buyer/End-User: Shenzhen LOHAS Supply Chain Management Co., Ltd.'], 'id': 'LohaCompanyltd_20191209_F-1_EX-10.16_11917878_EX-10.16_Supply Agreement__Parties'}]
>>> references = [{'answers': {'answer_start': [143, 49], 'text': ['The seller:', 'The buyer/End-User: Shenzhen LOHAS Supply Chain Management Co., Ltd.']}, 'id': 'LohaCompanyltd_20191209_F-1_EX-10.16_11917878_EX-10.16_Supply Agreement__Parties'}]
>>> cuad_metric = datasets.load_metric(\"cuad\")
>>> results = cuad_metric.compute(predictions=predictions, references=references)
>>> print(results)
{'exact_match': 100.0, 'f1': 100.0, 'aupr': 0.0, 'prec_at_80_recall': 1.0, 'prec_at_90_recall': 1.0}
"""
@datasets.utils.file_utils.add_start_docstrings(_DESCRIPTION , _KWARGS_DESCRIPTION )
class _a ( datasets.Metric ):
"""simple docstring"""
def _UpperCAmelCase ( self ) -> Optional[int]:
return datasets.MetricInfo(
description=_DESCRIPTION , citation=_CITATION , inputs_description=_KWARGS_DESCRIPTION , features=datasets.Features(
{
'predictions': {
'id': datasets.Value('string' ),
'prediction_text': datasets.features.Sequence(datasets.Value('string' ) ),
},
'references': {
'id': datasets.Value('string' ),
'answers': datasets.features.Sequence(
{
'text': datasets.Value('string' ),
'answer_start': datasets.Value('int32' ),
} ),
},
} ) , codebase_urls=['https://www.atticusprojectai.org/cuad'] , reference_urls=['https://www.atticusprojectai.org/cuad'] , )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> Dict:
UpperCamelCase_ = {prediction['id']: prediction['prediction_text'] for prediction in predictions}
UpperCamelCase_ = [
{
'paragraphs': [
{
'qas': [
{
'answers': [{'text': answer_text} for answer_text in ref['answers']['text']],
'id': ref['id'],
}
for ref in references
]
}
]
}
]
UpperCamelCase_ = evaluate(dataset=_UpperCAmelCase , predictions=_UpperCAmelCase )
return score
| 23 | 0 |
from typing import TYPE_CHECKING
from ...utils import OptionalDependencyNotAvailable, _LazyModule, is_torch_available
lowerCAmelCase_ = {'''configuration_sew''': ['''SEW_PRETRAINED_CONFIG_ARCHIVE_MAP''', '''SEWConfig''']}
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
lowerCAmelCase_ = [
'''SEW_PRETRAINED_MODEL_ARCHIVE_LIST''',
'''SEWForCTC''',
'''SEWForSequenceClassification''',
'''SEWModel''',
'''SEWPreTrainedModel''',
]
if TYPE_CHECKING:
from .configuration_sew import SEW_PRETRAINED_CONFIG_ARCHIVE_MAP, SEWConfig
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_sew import (
SEW_PRETRAINED_MODEL_ARCHIVE_LIST,
SEWForCTC,
SEWForSequenceClassification,
SEWModel,
SEWPreTrainedModel,
)
else:
import sys
lowerCAmelCase_ = _LazyModule(__name__, globals()['''__file__'''], _import_structure, module_spec=__spec__) | 39 |
import os
import tempfile
from functools import partial
from unittest import TestCase
from unittest.mock import patch
import datasets
import datasets.config
from .utils import require_beam
class _a ( datasets.BeamBasedBuilder ):
"""simple docstring"""
def _UpperCAmelCase ( self ) -> List[str]:
return datasets.DatasetInfo(
features=datasets.Features({'content': datasets.Value('string' )} ) , supervised_keys=_UpperCAmelCase , )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> Optional[Any]:
return [datasets.SplitGenerator(name=datasets.Split.TRAIN , gen_kwargs={'examples': get_test_dummy_examples()} )]
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> Tuple:
import apache_beam as beam
return pipeline | "Load Examples" >> beam.Create(_UpperCAmelCase )
class _a ( datasets.BeamBasedBuilder ):
"""simple docstring"""
def _UpperCAmelCase ( self ) -> Any:
return datasets.DatasetInfo(
features=datasets.Features({'a': datasets.Sequence({'b': datasets.Value('string' )} )} ) , supervised_keys=_UpperCAmelCase , )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> Tuple:
return [
datasets.SplitGenerator(name=datasets.Split.TRAIN , gen_kwargs={'examples': get_test_nested_examples()} )
]
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> Union[str, Any]:
import apache_beam as beam
return pipeline | "Load Examples" >> beam.Create(_UpperCAmelCase )
def _snake_case ():
return [(i, {"content": content}) for i, content in enumerate(['foo', 'bar', 'foobar'])]
def _snake_case ():
return [(i, {"a": {"b": [content]}}) for i, content in enumerate(['foo', 'bar', 'foobar'])]
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
@require_beam
def _UpperCAmelCase ( self ) -> Dict:
UpperCamelCase_ = len(get_test_dummy_examples() )
with tempfile.TemporaryDirectory() as tmp_cache_dir:
UpperCamelCase_ = DummyBeamDataset(cache_dir=_UpperCAmelCase , beam_runner='DirectRunner' )
builder.download_and_prepare()
self.assertTrue(
os.path.exists(
os.path.join(_UpperCAmelCase , builder.name , 'default' , '0.0.0' , f"""{builder.name}-train.arrow""" ) ) )
self.assertDictEqual(builder.info.features , datasets.Features({'content': datasets.Value('string' )} ) )
UpperCamelCase_ = builder.as_dataset()
self.assertEqual(dset['train'].num_rows , _UpperCAmelCase )
self.assertEqual(dset['train'].info.splits['train'].num_examples , _UpperCAmelCase )
self.assertDictEqual(dset['train'][0] , get_test_dummy_examples()[0][1] )
self.assertDictEqual(
dset['train'][expected_num_examples - 1] , get_test_dummy_examples()[expected_num_examples - 1][1] )
self.assertTrue(
os.path.exists(os.path.join(_UpperCAmelCase , builder.name , 'default' , '0.0.0' , 'dataset_info.json' ) ) )
del dset
@require_beam
def _UpperCAmelCase ( self ) -> List[str]:
import apache_beam as beam
UpperCamelCase_ = beam.io.parquetio.WriteToParquet
UpperCamelCase_ = len(get_test_dummy_examples() )
with tempfile.TemporaryDirectory() as tmp_cache_dir:
UpperCamelCase_ = DummyBeamDataset(cache_dir=_UpperCAmelCase , beam_runner='DirectRunner' )
with patch('apache_beam.io.parquetio.WriteToParquet' ) as write_parquet_mock:
UpperCamelCase_ = partial(_UpperCAmelCase , num_shards=2 )
builder.download_and_prepare()
self.assertTrue(
os.path.exists(
os.path.join(
_UpperCAmelCase , builder.name , 'default' , '0.0.0' , f"""{builder.name}-train-00000-of-00002.arrow""" ) ) )
self.assertTrue(
os.path.exists(
os.path.join(
_UpperCAmelCase , builder.name , 'default' , '0.0.0' , f"""{builder.name}-train-00000-of-00002.arrow""" ) ) )
self.assertDictEqual(builder.info.features , datasets.Features({'content': datasets.Value('string' )} ) )
UpperCamelCase_ = builder.as_dataset()
self.assertEqual(dset['train'].num_rows , _UpperCAmelCase )
self.assertEqual(dset['train'].info.splits['train'].num_examples , _UpperCAmelCase )
# Order is not preserved when sharding, so we just check that all the elements are there
self.assertListEqual(sorted(dset['train']['content'] ) , sorted(['foo', 'bar', 'foobar'] ) )
self.assertTrue(
os.path.exists(os.path.join(_UpperCAmelCase , builder.name , 'default' , '0.0.0' , 'dataset_info.json' ) ) )
del dset
@require_beam
def _UpperCAmelCase ( self ) -> Any:
with tempfile.TemporaryDirectory() as tmp_cache_dir:
UpperCamelCase_ = DummyBeamDataset(cache_dir=_UpperCAmelCase )
self.assertRaises(datasets.builder.MissingBeamOptions , builder.download_and_prepare )
@require_beam
def _UpperCAmelCase ( self ) -> Union[str, Any]:
UpperCamelCase_ = len(get_test_nested_examples() )
with tempfile.TemporaryDirectory() as tmp_cache_dir:
UpperCamelCase_ = NestedBeamDataset(cache_dir=_UpperCAmelCase , beam_runner='DirectRunner' )
builder.download_and_prepare()
self.assertTrue(
os.path.exists(
os.path.join(_UpperCAmelCase , builder.name , 'default' , '0.0.0' , f"""{builder.name}-train.arrow""" ) ) )
self.assertDictEqual(
builder.info.features , datasets.Features({'a': datasets.Sequence({'b': datasets.Value('string' )} )} ) )
UpperCamelCase_ = builder.as_dataset()
self.assertEqual(dset['train'].num_rows , _UpperCAmelCase )
self.assertEqual(dset['train'].info.splits['train'].num_examples , _UpperCAmelCase )
self.assertDictEqual(dset['train'][0] , get_test_nested_examples()[0][1] )
self.assertDictEqual(
dset['train'][expected_num_examples - 1] , get_test_nested_examples()[expected_num_examples - 1][1] )
self.assertTrue(
os.path.exists(os.path.join(_UpperCAmelCase , builder.name , 'default' , '0.0.0' , 'dataset_info.json' ) ) )
del dset
| 23 | 0 |
import itertools
import random
import unittest
import numpy as np
from transformers import WAV_2_VEC_2_PRETRAINED_MODEL_ARCHIVE_LIST, WavaVecaConfig, WavaVecaFeatureExtractor
from transformers.testing_utils import require_torch, slow
from ...test_sequence_feature_extraction_common import SequenceFeatureExtractionTestMixin
__UpperCAmelCase = random.Random()
def UpperCamelCase ( snake_case__ : List[Any] , snake_case__ : str=1.0 , snake_case__ : int=None , snake_case__ : Union[str, Any]=None ) -> Any:
if rng is None:
UpperCamelCase : int = global_rng
UpperCamelCase : Union[str, Any] = []
for batch_idx in range(shape[0] ):
values.append([] )
for _ in range(shape[1] ):
values[-1].append(rng.random() * scale )
return values
class lowerCAmelCase_ ( unittest.TestCase ):
def __init__( self, SCREAMING_SNAKE_CASE_, SCREAMING_SNAKE_CASE_=7, SCREAMING_SNAKE_CASE_=400, SCREAMING_SNAKE_CASE_=2000, SCREAMING_SNAKE_CASE_=1, SCREAMING_SNAKE_CASE_=0.0, SCREAMING_SNAKE_CASE_=1_6000, SCREAMING_SNAKE_CASE_=True, SCREAMING_SNAKE_CASE_=True, ) -> List[str]:
UpperCamelCase : Dict = parent
UpperCamelCase : Dict = batch_size
UpperCamelCase : Any = min_seq_length
UpperCamelCase : Optional[int] = max_seq_length
UpperCamelCase : Optional[int] = (self.max_seq_length - self.min_seq_length) // (self.batch_size - 1)
UpperCamelCase : Tuple = feature_size
UpperCamelCase : Any = padding_value
UpperCamelCase : Tuple = sampling_rate
UpperCamelCase : Optional[Any] = return_attention_mask
UpperCamelCase : Optional[Any] = do_normalize
def snake_case_ ( self ) -> Union[str, Any]:
return {
"feature_size": self.feature_size,
"padding_value": self.padding_value,
"sampling_rate": self.sampling_rate,
"return_attention_mask": self.return_attention_mask,
"do_normalize": self.do_normalize,
}
def snake_case_ ( self, SCREAMING_SNAKE_CASE_=False, SCREAMING_SNAKE_CASE_=False ) -> Union[str, Any]:
def _flatten(SCREAMING_SNAKE_CASE_ ):
return list(itertools.chain(*SCREAMING_SNAKE_CASE_ ) )
if equal_length:
UpperCamelCase : List[str] = floats_list((self.batch_size, self.max_seq_length) )
else:
# make sure that inputs increase in size
UpperCamelCase : Union[str, Any] = [
_flatten(floats_list((x, self.feature_size) ) )
for x in range(self.min_seq_length, self.max_seq_length, self.seq_length_diff )
]
if numpify:
UpperCamelCase : str = [np.asarray(SCREAMING_SNAKE_CASE_ ) for x in speech_inputs]
return speech_inputs
class lowerCAmelCase_ ( a__ , unittest.TestCase ):
UpperCAmelCase__ : Any = WavaVecaFeatureExtractor
def snake_case_ ( self ) -> Union[str, Any]:
UpperCamelCase : Tuple = WavaVecaFeatureExtractionTester(self )
def snake_case_ ( self, SCREAMING_SNAKE_CASE_ ) -> Optional[int]:
self.assertTrue(np.all(np.mean(SCREAMING_SNAKE_CASE_, axis=0 ) < 1e-3 ) )
self.assertTrue(np.all(np.abs(np.var(SCREAMING_SNAKE_CASE_, axis=0 ) - 1 ) < 1e-3 ) )
def snake_case_ ( self ) -> Optional[int]:
# Tests that all call wrap to encode_plus and batch_encode_plus
UpperCamelCase : Tuple = self.feature_extraction_class(**self.feat_extract_tester.prepare_feat_extract_dict() )
# create three inputs of length 800, 1000, and 1200
UpperCamelCase : Any = [floats_list((1, x) )[0] for x in range(800, 1400, 200 )]
UpperCamelCase : Dict = [np.asarray(SCREAMING_SNAKE_CASE_ ) for speech_input in speech_inputs]
# Test not batched input
UpperCamelCase : List[Any] = feat_extract(speech_inputs[0], return_tensors='np' ).input_values
UpperCamelCase : Union[str, Any] = feat_extract(np_speech_inputs[0], return_tensors='np' ).input_values
self.assertTrue(np.allclose(SCREAMING_SNAKE_CASE_, SCREAMING_SNAKE_CASE_, atol=1e-3 ) )
# Test batched
UpperCamelCase : List[Any] = feat_extract(SCREAMING_SNAKE_CASE_, return_tensors='np' ).input_values
UpperCamelCase : int = feat_extract(SCREAMING_SNAKE_CASE_, return_tensors='np' ).input_values
for enc_seq_a, enc_seq_a in zip(SCREAMING_SNAKE_CASE_, SCREAMING_SNAKE_CASE_ ):
self.assertTrue(np.allclose(SCREAMING_SNAKE_CASE_, SCREAMING_SNAKE_CASE_, atol=1e-3 ) )
# Test 2-D numpy arrays are batched.
UpperCamelCase : Tuple = [floats_list((1, x) )[0] for x in (800, 800, 800)]
UpperCamelCase : Optional[int] = np.asarray(SCREAMING_SNAKE_CASE_ )
UpperCamelCase : Union[str, Any] = feat_extract(SCREAMING_SNAKE_CASE_, return_tensors='np' ).input_values
UpperCamelCase : Dict = feat_extract(SCREAMING_SNAKE_CASE_, return_tensors='np' ).input_values
for enc_seq_a, enc_seq_a in zip(SCREAMING_SNAKE_CASE_, SCREAMING_SNAKE_CASE_ ):
self.assertTrue(np.allclose(SCREAMING_SNAKE_CASE_, SCREAMING_SNAKE_CASE_, atol=1e-3 ) )
def snake_case_ ( self ) -> int:
UpperCamelCase : Dict = self.feature_extraction_class(**self.feat_extract_tester.prepare_feat_extract_dict() )
UpperCamelCase : Dict = [floats_list((1, x) )[0] for x in range(800, 1400, 200 )]
UpperCamelCase : str = ['longest', 'max_length', 'do_not_pad']
UpperCamelCase : Any = [None, 1600, None]
for max_length, padding in zip(SCREAMING_SNAKE_CASE_, SCREAMING_SNAKE_CASE_ ):
UpperCamelCase : Optional[Any] = feat_extract(SCREAMING_SNAKE_CASE_, padding=SCREAMING_SNAKE_CASE_, max_length=SCREAMING_SNAKE_CASE_, return_tensors='np' )
UpperCamelCase : Tuple = processed.input_values
self._check_zero_mean_unit_variance(input_values[0][:800] )
self.assertTrue(input_values[0][800:].sum() < 1e-6 )
self._check_zero_mean_unit_variance(input_values[1][:1000] )
self.assertTrue(input_values[0][1000:].sum() < 1e-6 )
self._check_zero_mean_unit_variance(input_values[2][:1200] )
def snake_case_ ( self ) -> Tuple:
UpperCamelCase : List[str] = self.feature_extraction_class(**self.feat_extract_tester.prepare_feat_extract_dict() )
UpperCamelCase : Tuple = range(800, 1400, 200 )
UpperCamelCase : str = [floats_list((1, x) )[0] for x in lengths]
UpperCamelCase : int = ['longest', 'max_length', 'do_not_pad']
UpperCamelCase : List[str] = [None, 1600, None]
for max_length, padding in zip(SCREAMING_SNAKE_CASE_, SCREAMING_SNAKE_CASE_ ):
UpperCamelCase : Tuple = feat_extract(SCREAMING_SNAKE_CASE_, max_length=SCREAMING_SNAKE_CASE_, padding=SCREAMING_SNAKE_CASE_ )
UpperCamelCase : Optional[int] = processed.input_values
self._check_zero_mean_unit_variance(input_values[0][:800] )
self._check_zero_mean_unit_variance(input_values[1][:1000] )
self._check_zero_mean_unit_variance(input_values[2][:1200] )
def snake_case_ ( self ) -> Optional[Any]:
UpperCamelCase : Optional[int] = self.feature_extraction_class(**self.feat_extract_tester.prepare_feat_extract_dict() )
UpperCamelCase : Optional[int] = [floats_list((1, x) )[0] for x in range(800, 1400, 200 )]
UpperCamelCase : int = feat_extract(
SCREAMING_SNAKE_CASE_, truncation=SCREAMING_SNAKE_CASE_, max_length=1000, padding='max_length', return_tensors='np' )
UpperCamelCase : Tuple = processed.input_values
self._check_zero_mean_unit_variance(input_values[0, :800] )
self._check_zero_mean_unit_variance(input_values[1] )
self._check_zero_mean_unit_variance(input_values[2] )
def snake_case_ ( self ) -> List[Any]:
UpperCamelCase : List[Any] = self.feature_extraction_class(**self.feat_extract_tester.prepare_feat_extract_dict() )
UpperCamelCase : Union[str, Any] = [floats_list((1, x) )[0] for x in range(800, 1400, 200 )]
UpperCamelCase : Any = feat_extract(
SCREAMING_SNAKE_CASE_, truncation=SCREAMING_SNAKE_CASE_, max_length=1000, padding='longest', return_tensors='np' )
UpperCamelCase : Dict = processed.input_values
self._check_zero_mean_unit_variance(input_values[0, :800] )
self._check_zero_mean_unit_variance(input_values[1, :1000] )
self._check_zero_mean_unit_variance(input_values[2] )
# make sure that if max_length < longest -> then pad to max_length
self.assertTrue(input_values.shape == (3, 1000) )
UpperCamelCase : str = [floats_list((1, x) )[0] for x in range(800, 1400, 200 )]
UpperCamelCase : Any = feat_extract(
SCREAMING_SNAKE_CASE_, truncation=SCREAMING_SNAKE_CASE_, max_length=2000, padding='longest', return_tensors='np' )
UpperCamelCase : int = processed.input_values
self._check_zero_mean_unit_variance(input_values[0, :800] )
self._check_zero_mean_unit_variance(input_values[1, :1000] )
self._check_zero_mean_unit_variance(input_values[2] )
# make sure that if max_length > longest -> then pad to longest
self.assertTrue(input_values.shape == (3, 1200) )
@require_torch
def snake_case_ ( self ) -> str:
import torch
UpperCamelCase : Any = self.feature_extraction_class(**self.feat_extract_tester.prepare_feat_extract_dict() )
UpperCamelCase : Dict = np.random.rand(100 ).astype(np.floataa )
UpperCamelCase : Dict = np_speech_inputs.tolist()
for inputs in [py_speech_inputs, np_speech_inputs]:
UpperCamelCase : Union[str, Any] = feature_extractor.pad([{'input_values': inputs}], return_tensors='np' )
self.assertTrue(np_processed.input_values.dtype == np.floataa )
UpperCamelCase : Any = feature_extractor.pad([{'input_values': inputs}], return_tensors='pt' )
self.assertTrue(pt_processed.input_values.dtype == torch.floataa )
@slow
@require_torch
def snake_case_ ( self ) -> Tuple:
# this test makes sure that models that are using
# group norm don't have their feature extractor return the
# attention_mask
for model_id in WAV_2_VEC_2_PRETRAINED_MODEL_ARCHIVE_LIST:
UpperCamelCase : int = WavaVecaConfig.from_pretrained(SCREAMING_SNAKE_CASE_ )
UpperCamelCase : Dict = WavaVecaFeatureExtractor.from_pretrained(SCREAMING_SNAKE_CASE_ )
# only "layer" feature extraction norm should make use of
# attention_mask
self.assertEqual(feat_extract.return_attention_mask, config.feat_extract_norm == 'layer' )
| 40 |
import argparse
import torch
from ...utils import logging
from . import AlbertConfig, AlbertForPreTraining, load_tf_weights_in_albert
logging.set_verbosity_info()
def _snake_case (__lowercase , __lowercase , __lowercase):
# Initialise PyTorch model
UpperCamelCase_ = AlbertConfig.from_json_file(__lowercase)
print(f"""Building PyTorch model from configuration: {config}""")
UpperCamelCase_ = AlbertForPreTraining(__lowercase)
# Load weights from tf checkpoint
load_tf_weights_in_albert(__lowercase , __lowercase , __lowercase)
# Save pytorch-model
print(f"""Save PyTorch model to {pytorch_dump_path}""")
torch.save(model.state_dict() , __lowercase)
if __name__ == "__main__":
snake_case__ : Tuple = argparse.ArgumentParser()
# Required parameters
parser.add_argument(
"""--tf_checkpoint_path""", default=None, type=str, required=True, help="""Path to the TensorFlow checkpoint path."""
)
parser.add_argument(
"""--albert_config_file""",
default=None,
type=str,
required=True,
help=(
"""The config json file corresponding to the pre-trained ALBERT model. \n"""
"""This specifies the model architecture."""
),
)
parser.add_argument(
"""--pytorch_dump_path""", default=None, type=str, required=True, help="""Path to the output PyTorch model."""
)
snake_case__ : str = parser.parse_args()
convert_tf_checkpoint_to_pytorch(args.tf_checkpoint_path, args.albert_config_file, args.pytorch_dump_path)
| 23 | 0 |
'''simple docstring'''
import math
from numpy import inf
from scipy.integrate import quad
def _A ( A__ ):
"""simple docstring"""
if num <= 0:
raise ValueError('''math domain error''' )
return quad(A__ , 0 , A__ , args=(A__) )[0]
def _A ( A__ , A__ ):
"""simple docstring"""
return math.pow(A__ , z - 1 ) * math.exp(-x )
if __name__ == "__main__":
from doctest import testmod
testmod()
| 41 |
from transformers import BertTokenizer, EncoderDecoderModel, SeqaSeqTrainer, SeqaSeqTrainingArguments
from transformers.testing_utils import TestCasePlus, require_torch, slow
from transformers.utils import is_datasets_available
if is_datasets_available():
import datasets
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
@slow
@require_torch
def _UpperCAmelCase ( self ) -> List[Any]:
UpperCamelCase_ = EncoderDecoderModel.from_encoder_decoder_pretrained('prajjwal1/bert-tiny' , 'prajjwal1/bert-tiny' )
UpperCamelCase_ = BertTokenizer.from_pretrained('bert-base-uncased' )
UpperCamelCase_ = bertabert.config.encoder.vocab_size
UpperCamelCase_ = tokenizer.sep_token_id
UpperCamelCase_ = tokenizer.cls_token_id
UpperCamelCase_ = 128
UpperCamelCase_ = datasets.load_dataset('cnn_dailymail' , '3.0.0' , split='train[:1%]' )
UpperCamelCase_ = datasets.load_dataset('cnn_dailymail' , '3.0.0' , split='validation[:1%]' )
UpperCamelCase_ = train_dataset.select(range(32 ) )
UpperCamelCase_ = val_dataset.select(range(16 ) )
UpperCamelCase_ = 4
def _map_to_encoder_decoder_inputs(_UpperCAmelCase ):
# Tokenizer will automatically set [BOS] <text> [EOS]
UpperCamelCase_ = tokenizer(batch['article'] , padding='max_length' , truncation=_UpperCAmelCase , max_length=512 )
UpperCamelCase_ = tokenizer(batch['highlights'] , padding='max_length' , truncation=_UpperCAmelCase , max_length=128 )
UpperCamelCase_ = inputs.input_ids
UpperCamelCase_ = inputs.attention_mask
UpperCamelCase_ = outputs.input_ids
UpperCamelCase_ = outputs.input_ids.copy()
UpperCamelCase_ = [
[-100 if token == tokenizer.pad_token_id else token for token in labels] for labels in batch['labels']
]
UpperCamelCase_ = outputs.attention_mask
assert all(len(_UpperCAmelCase ) == 512 for x in inputs.input_ids )
assert all(len(_UpperCAmelCase ) == 128 for x in outputs.input_ids )
return batch
def _compute_metrics(_UpperCAmelCase ):
UpperCamelCase_ = pred.label_ids
UpperCamelCase_ = pred.predictions
# all unnecessary tokens are removed
UpperCamelCase_ = tokenizer.batch_decode(_UpperCAmelCase , skip_special_tokens=_UpperCAmelCase )
UpperCamelCase_ = tokenizer.batch_decode(_UpperCAmelCase , skip_special_tokens=_UpperCAmelCase )
UpperCamelCase_ = sum([int(pred_str[i] == label_str[i] ) for i in range(len(_UpperCAmelCase ) )] ) / len(_UpperCAmelCase )
return {"accuracy": accuracy}
# map train dataset
UpperCamelCase_ = train_dataset.map(
_map_to_encoder_decoder_inputs , batched=_UpperCAmelCase , batch_size=_UpperCAmelCase , remove_columns=['article', 'highlights'] , )
train_dataset.set_format(
type='torch' , columns=['input_ids', 'attention_mask', 'decoder_input_ids', 'decoder_attention_mask', 'labels'] , )
# same for validation dataset
UpperCamelCase_ = val_dataset.map(
_map_to_encoder_decoder_inputs , batched=_UpperCAmelCase , batch_size=_UpperCAmelCase , remove_columns=['article', 'highlights'] , )
val_dataset.set_format(
type='torch' , columns=['input_ids', 'attention_mask', 'decoder_input_ids', 'decoder_attention_mask', 'labels'] , )
UpperCamelCase_ = self.get_auto_remove_tmp_dir()
UpperCamelCase_ = SeqaSeqTrainingArguments(
output_dir=_UpperCAmelCase , per_device_train_batch_size=_UpperCAmelCase , per_device_eval_batch_size=_UpperCAmelCase , predict_with_generate=_UpperCAmelCase , evaluation_strategy='steps' , do_train=_UpperCAmelCase , do_eval=_UpperCAmelCase , warmup_steps=0 , eval_steps=2 , logging_steps=2 , )
# instantiate trainer
UpperCamelCase_ = SeqaSeqTrainer(
model=_UpperCAmelCase , args=_UpperCAmelCase , compute_metrics=_compute_metrics , train_dataset=_UpperCAmelCase , eval_dataset=_UpperCAmelCase , tokenizer=_UpperCAmelCase , )
# start training
trainer.train()
| 23 | 0 |
'''simple docstring'''
from ..utils import is_flax_available, is_torch_available
if is_torch_available():
from .autoencoder_kl import AutoencoderKL
from .controlnet import ControlNetModel
from .dual_transformer_ad import DualTransformeraDModel
from .modeling_utils import ModelMixin
from .prior_transformer import PriorTransformer
from .ta_film_transformer import TaFilmDecoder
from .transformer_ad import TransformeraDModel
from .unet_ad import UNetaDModel
from .unet_ad import UNetaDModel
from .unet_ad_condition import UNetaDConditionModel
from .unet_ad_condition import UNetaDConditionModel
from .vq_model import VQModel
if is_flax_available():
from .controlnet_flax import FlaxControlNetModel
from .unet_ad_condition_flax import FlaxUNetaDConditionModel
from .vae_flax import FlaxAutoencoderKL
| 42 |
import argparse
import os
import evaluate
import torch
from datasets import load_dataset
from torch.optim import AdamW
from torch.utils.data import DataLoader
from transformers import AutoModelForSequenceClassification, AutoTokenizer, get_linear_schedule_with_warmup, set_seed
from accelerate import Accelerator, DistributedType
########################################################################
# This is a fully working simple example to use Accelerate
# and perform gradient accumulation
#
# This example trains a Bert base model on GLUE MRPC
# in any of the following settings (with the same script):
# - single CPU or single GPU
# - multi GPUS (using PyTorch distributed mode)
# - (multi) TPUs
# - fp16 (mixed-precision) or fp32 (normal precision)
#
# To run it in each of these various modes, follow the instructions
# in the readme for examples:
# https://github.com/huggingface/accelerate/tree/main/examples
#
########################################################################
snake_case__ : Dict = 1_6
snake_case__ : List[str] = 3_2
def _snake_case (__lowercase , __lowercase = 16):
UpperCamelCase_ = AutoTokenizer.from_pretrained('bert-base-cased')
UpperCamelCase_ = load_dataset('glue' , 'mrpc')
def tokenize_function(__lowercase):
# max_length=None => use the model max length (it's actually the default)
UpperCamelCase_ = tokenizer(examples['sentence1'] , examples['sentence2'] , truncation=__lowercase , max_length=__lowercase)
return outputs
# Apply the method we just defined to all the examples in all the splits of the dataset
# starting with the main process first:
with accelerator.main_process_first():
UpperCamelCase_ = datasets.map(
__lowercase , batched=__lowercase , remove_columns=['idx', 'sentence1', 'sentence2'] , )
# We also rename the 'label' column to 'labels' which is the expected name for labels by the models of the
# transformers library
UpperCamelCase_ = tokenized_datasets.rename_column('label' , 'labels')
def collate_fn(__lowercase):
# On TPU it's best to pad everything to the same length or training will be very slow.
UpperCamelCase_ = 128 if accelerator.distributed_type == DistributedType.TPU else None
# When using mixed precision we want round multiples of 8/16
if accelerator.mixed_precision == "fp8":
UpperCamelCase_ = 16
elif accelerator.mixed_precision != "no":
UpperCamelCase_ = 8
else:
UpperCamelCase_ = None
return tokenizer.pad(
__lowercase , padding='longest' , max_length=__lowercase , pad_to_multiple_of=__lowercase , return_tensors='pt' , )
# Instantiate dataloaders.
UpperCamelCase_ = DataLoader(
tokenized_datasets['train'] , shuffle=__lowercase , collate_fn=__lowercase , batch_size=__lowercase)
UpperCamelCase_ = DataLoader(
tokenized_datasets['validation'] , shuffle=__lowercase , collate_fn=__lowercase , batch_size=__lowercase)
return train_dataloader, eval_dataloader
# For testing only
if os.environ.get("""TESTING_MOCKED_DATALOADERS""", None) == "1":
from accelerate.test_utils.training import mocked_dataloaders
snake_case__ : List[str] = mocked_dataloaders # noqa: F811
def _snake_case (__lowercase , __lowercase):
# For testing only
if os.environ.get('TESTING_MOCKED_DATALOADERS' , __lowercase) == "1":
UpperCamelCase_ = 2
# New Code #
UpperCamelCase_ = int(args.gradient_accumulation_steps)
# Initialize accelerator
UpperCamelCase_ = Accelerator(
cpu=args.cpu , mixed_precision=args.mixed_precision , gradient_accumulation_steps=__lowercase)
if accelerator.distributed_type == DistributedType.TPU and gradient_accumulation_steps > 1:
raise NotImplementedError(
'Gradient accumulation on TPUs is currently not supported. Pass `gradient_accumulation_steps=1`')
# Sample hyper-parameters for learning rate, batch size, seed and a few other HPs
UpperCamelCase_ = config['lr']
UpperCamelCase_ = int(config['num_epochs'])
UpperCamelCase_ = int(config['seed'])
UpperCamelCase_ = int(config['batch_size'])
UpperCamelCase_ = evaluate.load('glue' , 'mrpc')
set_seed(__lowercase)
UpperCamelCase_ , UpperCamelCase_ = get_dataloaders(__lowercase , __lowercase)
# Instantiate the model (we build the model here so that the seed also control new weights initialization)
UpperCamelCase_ = AutoModelForSequenceClassification.from_pretrained('bert-base-cased' , return_dict=__lowercase)
# We could avoid this line since the accelerator is set with `device_placement=True` (default value).
# Note that if you are placing tensors on devices manually, this line absolutely needs to be before the optimizer
# creation otherwise training will not work on TPU (`accelerate` will kindly throw an error to make us aware of that).
UpperCamelCase_ = model.to(accelerator.device)
# Instantiate optimizer
UpperCamelCase_ = AdamW(params=model.parameters() , lr=__lowercase)
# Instantiate scheduler
UpperCamelCase_ = get_linear_schedule_with_warmup(
optimizer=__lowercase , num_warmup_steps=100 , num_training_steps=(len(__lowercase) * num_epochs) , )
# Prepare everything
# There is no specific order to remember, we just need to unpack the objects in the same order we gave them to the
# prepare method.
UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ = accelerator.prepare(
__lowercase , __lowercase , __lowercase , __lowercase , __lowercase)
# Now we train the model
for epoch in range(__lowercase):
model.train()
for step, batch in enumerate(__lowercase):
# We could avoid this line since we set the accelerator with `device_placement=True`.
batch.to(accelerator.device)
# New code #
# We use the new `accumulate` context manager to perform gradient accumulation
# We also currently do not support TPUs nor advise it as bugs were found on the XLA side when running our tests.
with accelerator.accumulate(__lowercase):
UpperCamelCase_ = model(**__lowercase)
UpperCamelCase_ = output.loss
accelerator.backward(__lowercase)
optimizer.step()
lr_scheduler.step()
optimizer.zero_grad()
model.eval()
for step, batch in enumerate(__lowercase):
# We could avoid this line since we set the accelerator with `device_placement=True`.
batch.to(accelerator.device)
with torch.no_grad():
UpperCamelCase_ = model(**__lowercase)
UpperCamelCase_ = outputs.logits.argmax(dim=-1)
UpperCamelCase_ , UpperCamelCase_ = accelerator.gather_for_metrics((predictions, batch['labels']))
metric.add_batch(
predictions=__lowercase , references=__lowercase , )
UpperCamelCase_ = metric.compute()
# Use accelerator.print to print only on the main process.
accelerator.print(f"""epoch {epoch}:""" , __lowercase)
def _snake_case ():
UpperCamelCase_ = argparse.ArgumentParser(description='Simple example of training script.')
parser.add_argument(
'--mixed_precision' , type=__lowercase , default=__lowercase , choices=['no', 'fp16', 'bf16', 'fp8'] , help='Whether to use mixed precision. Choose'
'between fp16 and bf16 (bfloat16). Bf16 requires PyTorch >= 1.10.'
'and an Nvidia Ampere GPU.' , )
# New Code #
parser.add_argument(
'--gradient_accumulation_steps' , type=__lowercase , default=1 , help='The number of minibatches to be ran before gradients are accumulated.' , )
parser.add_argument('--cpu' , action='store_true' , help='If passed, will train on the CPU.')
UpperCamelCase_ = parser.parse_args()
UpperCamelCase_ = {'lr': 2e-5, 'num_epochs': 3, 'seed': 42, 'batch_size': 16}
training_function(__lowercase , __lowercase)
if __name__ == "__main__":
main()
| 23 | 0 |
from typing import TYPE_CHECKING
from ...utils import OptionalDependencyNotAvailable, _LazyModule, is_sentencepiece_available
lowerCAmelCase = {}
try:
if not is_sentencepiece_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
lowerCAmelCase = ['BartphoTokenizer']
if TYPE_CHECKING:
try:
if not is_sentencepiece_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .tokenization_bartpho import BartphoTokenizer
else:
import sys
lowerCAmelCase = _LazyModule(__name__, globals()['__file__'], _import_structure, module_spec=__spec__)
| 43 |
import inspect
import unittest
import numpy as np
from tests.test_modeling_common import floats_tensor
from transformers import MaskaFormerConfig, is_torch_available, is_vision_available
from transformers.testing_utils import require_torch, require_torch_multi_gpu, require_vision, slow, torch_device
from transformers.utils import cached_property
from ...test_configuration_common import ConfigTester
from ...test_modeling_common import ModelTesterMixin
from ...test_pipeline_mixin import PipelineTesterMixin
if is_torch_available():
import torch
from transformers import MaskaFormerForUniversalSegmentation, MaskaFormerModel
if is_vision_available():
from transformers import MaskaFormerImageProcessor
if is_vision_available():
from PIL import Image
class _a :
"""simple docstring"""
def __init__( self , _UpperCAmelCase , _UpperCAmelCase=2 , _UpperCAmelCase=True , _UpperCAmelCase=False , _UpperCAmelCase=10 , _UpperCAmelCase=3 , _UpperCAmelCase=32 * 8 , _UpperCAmelCase=32 * 8 , _UpperCAmelCase=4 , _UpperCAmelCase=64 , ) -> List[Any]:
UpperCamelCase_ = parent
UpperCamelCase_ = batch_size
UpperCamelCase_ = is_training
UpperCamelCase_ = use_auxiliary_loss
UpperCamelCase_ = num_queries
UpperCamelCase_ = num_channels
UpperCamelCase_ = min_size
UpperCamelCase_ = max_size
UpperCamelCase_ = num_labels
UpperCamelCase_ = hidden_dim
UpperCamelCase_ = hidden_dim
def _UpperCAmelCase ( self ) -> List[str]:
UpperCamelCase_ = floats_tensor([self.batch_size, self.num_channels, self.min_size, self.max_size] ).to(
_UpperCAmelCase )
UpperCamelCase_ = torch.ones([self.batch_size, self.min_size, self.max_size] , device=_UpperCAmelCase )
UpperCamelCase_ = (
torch.rand([self.batch_size, self.num_labels, self.min_size, self.max_size] , device=_UpperCAmelCase ) > 0.5
).float()
UpperCamelCase_ = (torch.rand((self.batch_size, self.num_labels) , device=_UpperCAmelCase ) > 0.5).long()
UpperCamelCase_ = self.get_config()
return config, pixel_values, pixel_mask, mask_labels, class_labels
def _UpperCAmelCase ( self ) -> Optional[int]:
UpperCamelCase_ = MaskaFormerConfig(
hidden_size=self.hidden_dim , )
UpperCamelCase_ = self.num_queries
UpperCamelCase_ = self.num_labels
UpperCamelCase_ = [1, 1, 1, 1]
UpperCamelCase_ = self.num_channels
UpperCamelCase_ = 64
UpperCamelCase_ = 128
UpperCamelCase_ = self.hidden_dim
UpperCamelCase_ = self.hidden_dim
UpperCamelCase_ = self.hidden_dim
return config
def _UpperCAmelCase ( self ) -> Dict:
UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ = self.prepare_config_and_inputs()
UpperCamelCase_ = {'pixel_values': pixel_values, 'pixel_mask': pixel_mask}
return config, inputs_dict
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> List[Any]:
UpperCamelCase_ = output.encoder_hidden_states
UpperCamelCase_ = output.pixel_decoder_hidden_states
UpperCamelCase_ = output.transformer_decoder_hidden_states
self.parent.assertTrue(len(_UpperCAmelCase ) , len(config.backbone_config.depths ) )
self.parent.assertTrue(len(_UpperCAmelCase ) , len(config.backbone_config.depths ) )
self.parent.assertTrue(len(_UpperCAmelCase ) , config.decoder_layers )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase=False ) -> Any:
with torch.no_grad():
UpperCamelCase_ = MaskaFormerModel(config=_UpperCAmelCase )
model.to(_UpperCAmelCase )
model.eval()
UpperCamelCase_ = model(pixel_values=_UpperCAmelCase , pixel_mask=_UpperCAmelCase )
UpperCamelCase_ = model(_UpperCAmelCase , output_hidden_states=_UpperCAmelCase )
self.parent.assertEqual(
output.transformer_decoder_last_hidden_state.shape , (self.batch_size, self.num_queries, self.hidden_dim) , )
# let's ensure the other two hidden state exists
self.parent.assertTrue(output.pixel_decoder_last_hidden_state is not None )
self.parent.assertTrue(output.encoder_last_hidden_state is not None )
if output_hidden_states:
self.check_output_hidden_state(_UpperCAmelCase , _UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase ) -> Union[str, Any]:
UpperCamelCase_ = MaskaFormerForUniversalSegmentation(config=_UpperCAmelCase )
model.to(_UpperCAmelCase )
model.eval()
def comm_check_on_output(_UpperCAmelCase ):
# let's still check that all the required stuff is there
self.parent.assertTrue(result.transformer_decoder_last_hidden_state is not None )
self.parent.assertTrue(result.pixel_decoder_last_hidden_state is not None )
self.parent.assertTrue(result.encoder_last_hidden_state is not None )
# okay, now we need to check the logits shape
# due to the encoder compression, masks have a //4 spatial size
self.parent.assertEqual(
result.masks_queries_logits.shape , (self.batch_size, self.num_queries, self.min_size // 4, self.max_size // 4) , )
# + 1 for null class
self.parent.assertEqual(
result.class_queries_logits.shape , (self.batch_size, self.num_queries, self.num_labels + 1) )
with torch.no_grad():
UpperCamelCase_ = model(pixel_values=_UpperCAmelCase , pixel_mask=_UpperCAmelCase )
UpperCamelCase_ = model(_UpperCAmelCase )
comm_check_on_output(_UpperCAmelCase )
UpperCamelCase_ = model(
pixel_values=_UpperCAmelCase , pixel_mask=_UpperCAmelCase , mask_labels=_UpperCAmelCase , class_labels=_UpperCAmelCase )
comm_check_on_output(_UpperCAmelCase )
self.parent.assertTrue(result.loss is not None )
self.parent.assertEqual(result.loss.shape , torch.Size([1] ) )
@require_torch
class _a ( UpperCAmelCase__ , UpperCAmelCase__ , unittest.TestCase ):
"""simple docstring"""
A_ = (MaskaFormerModel, MaskaFormerForUniversalSegmentation) if is_torch_available() else ()
A_ = {"""feature-extraction""": MaskaFormerModel} if is_torch_available() else {}
A_ = False
A_ = False
A_ = False
A_ = False
def _UpperCAmelCase ( self ) -> Optional[Any]:
UpperCamelCase_ = MaskaFormerModelTester(self )
UpperCamelCase_ = ConfigTester(self , config_class=_UpperCAmelCase , has_text_modality=_UpperCAmelCase )
def _UpperCAmelCase ( self ) -> Union[str, Any]:
self.config_tester.run_common_tests()
def _UpperCAmelCase ( self ) -> Union[str, Any]:
UpperCamelCase_ , UpperCamelCase_ = self.model_tester.prepare_config_and_inputs_for_common()
self.model_tester.create_and_check_maskaformer_model(_UpperCAmelCase , **_UpperCAmelCase , output_hidden_states=_UpperCAmelCase )
def _UpperCAmelCase ( self ) -> List[Any]:
UpperCamelCase_ = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_maskaformer_instance_segmentation_head_model(*_UpperCAmelCase )
@unittest.skip(reason='Mask2Former does not use inputs_embeds' )
def _UpperCAmelCase ( self ) -> Any:
pass
@unittest.skip(reason='Mask2Former does not have a get_input_embeddings method' )
def _UpperCAmelCase ( self ) -> Optional[int]:
pass
@unittest.skip(reason='Mask2Former is not a generative model' )
def _UpperCAmelCase ( self ) -> Any:
pass
@unittest.skip(reason='Mask2Former does not use token embeddings' )
def _UpperCAmelCase ( self ) -> Optional[Any]:
pass
@require_torch_multi_gpu
@unittest.skip(
reason='Mask2Former has some layers using `add_module` which doesn\'t work well with `nn.DataParallel`' )
def _UpperCAmelCase ( self ) -> int:
pass
@unittest.skip('Will be fixed soon by reducing the size of the model used for common tests.' )
def _UpperCAmelCase ( self ) -> str:
pass
def _UpperCAmelCase ( self ) -> Optional[int]:
UpperCamelCase_ , UpperCamelCase_ = self.model_tester.prepare_config_and_inputs_for_common()
for model_class in self.all_model_classes:
UpperCamelCase_ = model_class(_UpperCAmelCase )
UpperCamelCase_ = inspect.signature(model.forward )
# signature.parameters is an OrderedDict => so arg_names order is deterministic
UpperCamelCase_ = [*signature.parameters.keys()]
UpperCamelCase_ = ['pixel_values']
self.assertListEqual(arg_names[:1] , _UpperCAmelCase )
@slow
def _UpperCAmelCase ( self ) -> Tuple:
for model_name in ["facebook/mask2former-swin-small-coco-instance"]:
UpperCamelCase_ = MaskaFormerModel.from_pretrained(_UpperCAmelCase )
self.assertIsNotNone(_UpperCAmelCase )
def _UpperCAmelCase ( self ) -> Dict:
UpperCamelCase_ = (self.model_tester.min_size,) * 2
UpperCamelCase_ = {
'pixel_values': torch.randn((2, 3, *size) , device=_UpperCAmelCase ),
'mask_labels': torch.randn((2, 10, *size) , device=_UpperCAmelCase ),
'class_labels': torch.zeros(2 , 10 , device=_UpperCAmelCase ).long(),
}
UpperCamelCase_ = self.model_tester.get_config()
UpperCamelCase_ = MaskaFormerForUniversalSegmentation(_UpperCAmelCase ).to(_UpperCAmelCase )
UpperCamelCase_ = model(**_UpperCAmelCase )
self.assertTrue(outputs.loss is not None )
def _UpperCAmelCase ( self ) -> str:
UpperCamelCase_ , UpperCamelCase_ = self.model_tester.prepare_config_and_inputs_for_common()
self.model_tester.create_and_check_maskaformer_model(_UpperCAmelCase , **_UpperCAmelCase , output_hidden_states=_UpperCAmelCase )
def _UpperCAmelCase ( self ) -> Optional[int]:
UpperCamelCase_ , UpperCamelCase_ = self.model_tester.prepare_config_and_inputs_for_common()
for model_class in self.all_model_classes:
UpperCamelCase_ = model_class(_UpperCAmelCase ).to(_UpperCAmelCase )
UpperCamelCase_ = model(**_UpperCAmelCase , output_attentions=_UpperCAmelCase )
self.assertTrue(outputs.attentions is not None )
def _UpperCAmelCase ( self ) -> List[Any]:
if not self.model_tester.is_training:
return
UpperCamelCase_ = self.all_model_classes[1]
UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ = self.model_tester.prepare_config_and_inputs()
UpperCamelCase_ = model_class(_UpperCAmelCase )
model.to(_UpperCAmelCase )
model.train()
UpperCamelCase_ = model(_UpperCAmelCase , mask_labels=_UpperCAmelCase , class_labels=_UpperCAmelCase ).loss
loss.backward()
def _UpperCAmelCase ( self ) -> int:
UpperCamelCase_ = self.all_model_classes[1]
UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ = self.model_tester.prepare_config_and_inputs()
UpperCamelCase_ = True
UpperCamelCase_ = True
UpperCamelCase_ = model_class(_UpperCAmelCase ).to(_UpperCAmelCase )
model.train()
UpperCamelCase_ = model(_UpperCAmelCase , mask_labels=_UpperCAmelCase , class_labels=_UpperCAmelCase )
UpperCamelCase_ = outputs.encoder_hidden_states[0]
encoder_hidden_states.retain_grad()
UpperCamelCase_ = outputs.pixel_decoder_hidden_states[0]
pixel_decoder_hidden_states.retain_grad()
UpperCamelCase_ = outputs.transformer_decoder_hidden_states[0]
transformer_decoder_hidden_states.retain_grad()
UpperCamelCase_ = outputs.attentions[0]
attentions.retain_grad()
outputs.loss.backward(retain_graph=_UpperCAmelCase )
self.assertIsNotNone(encoder_hidden_states.grad )
self.assertIsNotNone(pixel_decoder_hidden_states.grad )
self.assertIsNotNone(transformer_decoder_hidden_states.grad )
self.assertIsNotNone(attentions.grad )
snake_case__ : List[Any] = 1E-4
def _snake_case ():
UpperCamelCase_ = Image.open('./tests/fixtures/tests_samples/COCO/000000039769.png')
return image
@require_vision
@slow
class _a ( unittest.TestCase ):
"""simple docstring"""
@cached_property
def _UpperCAmelCase ( self ) -> Optional[int]:
return "facebook/mask2former-swin-small-coco-instance"
@cached_property
def _UpperCAmelCase ( self ) -> List[str]:
return MaskaFormerImageProcessor.from_pretrained(self.model_checkpoints ) if is_vision_available() else None
def _UpperCAmelCase ( self ) -> str:
UpperCamelCase_ = MaskaFormerModel.from_pretrained(self.model_checkpoints ).to(_UpperCAmelCase )
UpperCamelCase_ = self.default_image_processor
UpperCamelCase_ = prepare_img()
UpperCamelCase_ = image_processor(_UpperCAmelCase , return_tensors='pt' ).to(_UpperCAmelCase )
UpperCamelCase_ = inputs['pixel_values'].shape
# check size is divisible by 32
self.assertTrue((inputs_shape[-1] % 32) == 0 and (inputs_shape[-2] % 32) == 0 )
# check size
self.assertEqual(_UpperCAmelCase , (1, 3, 384, 384) )
with torch.no_grad():
UpperCamelCase_ = model(**_UpperCAmelCase )
UpperCamelCase_ = torch.tensor(
[[-0.2_7_9_0, -1.0_7_1_7, -1.1_6_6_8], [-0.5_1_2_8, -0.3_1_2_8, -0.4_9_8_7], [-0.5_8_3_2, 0.1_9_7_1, -0.0_1_9_7]] ).to(_UpperCAmelCase )
self.assertTrue(
torch.allclose(
outputs.encoder_last_hidden_state[0, 0, :3, :3] , _UpperCAmelCase , atol=_UpperCAmelCase ) )
UpperCamelCase_ = torch.tensor(
[[0.8_9_7_3, 1.1_8_4_7, 1.1_7_7_6], [1.1_9_3_4, 1.5_0_4_0, 1.5_1_2_8], [1.1_1_5_3, 1.4_4_8_6, 1.4_9_5_1]] ).to(_UpperCAmelCase )
self.assertTrue(
torch.allclose(
outputs.pixel_decoder_last_hidden_state[0, 0, :3, :3] , _UpperCAmelCase , atol=_UpperCAmelCase ) )
UpperCamelCase_ = torch.tensor(
[[2.1_1_5_2, 1.7_0_0_0, -0.8_6_0_3], [1.5_8_0_8, 1.8_0_0_4, -0.9_3_5_3], [1.6_0_4_3, 1.7_4_9_5, -0.5_9_9_9]] ).to(_UpperCAmelCase )
self.assertTrue(
torch.allclose(
outputs.transformer_decoder_last_hidden_state[0, :3, :3] , _UpperCAmelCase , atol=_UpperCAmelCase ) )
def _UpperCAmelCase ( self ) -> Optional[Any]:
UpperCamelCase_ = MaskaFormerForUniversalSegmentation.from_pretrained(self.model_checkpoints ).to(_UpperCAmelCase ).eval()
UpperCamelCase_ = self.default_image_processor
UpperCamelCase_ = prepare_img()
UpperCamelCase_ = image_processor(_UpperCAmelCase , return_tensors='pt' ).to(_UpperCAmelCase )
UpperCamelCase_ = inputs['pixel_values'].shape
# check size is divisible by 32
self.assertTrue((inputs_shape[-1] % 32) == 0 and (inputs_shape[-2] % 32) == 0 )
# check size
self.assertEqual(_UpperCAmelCase , (1, 3, 384, 384) )
with torch.no_grad():
UpperCamelCase_ = model(**_UpperCAmelCase )
# masks_queries_logits
UpperCamelCase_ = outputs.masks_queries_logits
self.assertEqual(
masks_queries_logits.shape , (1, model.config.num_queries, inputs_shape[-2] // 4, inputs_shape[-1] // 4) )
UpperCamelCase_ = [
[-8.7_8_3_9, -9.0_0_5_6, -8.8_1_2_1],
[-7.4_1_0_4, -7.0_3_1_3, -6.5_4_0_1],
[-6.6_1_0_5, -6.3_4_2_7, -6.4_6_7_5],
]
UpperCamelCase_ = torch.tensor(_UpperCAmelCase ).to(_UpperCAmelCase )
self.assertTrue(torch.allclose(masks_queries_logits[0, 0, :3, :3] , _UpperCAmelCase , atol=_UpperCAmelCase ) )
# class_queries_logits
UpperCamelCase_ = outputs.class_queries_logits
self.assertEqual(class_queries_logits.shape , (1, model.config.num_queries, model.config.num_labels + 1) )
UpperCamelCase_ = torch.tensor(
[
[1.8_3_2_4, -8.0_8_3_5, -4.1_9_2_2],
[0.8_4_5_0, -9.0_0_5_0, -3.6_0_5_3],
[0.3_0_4_5, -7.7_2_9_3, -3.0_2_7_5],
] ).to(_UpperCAmelCase )
self.assertTrue(torch.allclose(outputs.class_queries_logits[0, :3, :3] , _UpperCAmelCase , atol=_UpperCAmelCase ) )
def _UpperCAmelCase ( self ) -> Dict:
UpperCamelCase_ = MaskaFormerForUniversalSegmentation.from_pretrained(self.model_checkpoints ).to(_UpperCAmelCase ).eval()
UpperCamelCase_ = self.default_image_processor
UpperCamelCase_ = image_processor(
[np.zeros((3, 800, 1333) ), np.zeros((3, 800, 1333) )] , segmentation_maps=[np.zeros((384, 384) ).astype(np.floataa ), np.zeros((384, 384) ).astype(np.floataa )] , return_tensors='pt' , )
UpperCamelCase_ = inputs['pixel_values'].to(_UpperCAmelCase )
UpperCamelCase_ = [el.to(_UpperCAmelCase ) for el in inputs['mask_labels']]
UpperCamelCase_ = [el.to(_UpperCAmelCase ) for el in inputs['class_labels']]
with torch.no_grad():
UpperCamelCase_ = model(**_UpperCAmelCase )
self.assertTrue(outputs.loss is not None )
| 23 | 0 |
'''simple docstring'''
import math
import sys
def A_ ( _lowerCAmelCase : str ):
"""simple docstring"""
_lowerCamelCase : Dict = ""
try:
with open(_lowerCAmelCase , "rb" ) as binary_file:
_lowerCamelCase : List[Any] = binary_file.read()
for dat in data:
_lowerCamelCase : Optional[Any] = F'{dat:08b}'
result += curr_byte
return result
except OSError:
print("File not accessible" )
sys.exit()
def A_ ( _lowerCAmelCase : str ):
"""simple docstring"""
_lowerCamelCase : Optional[int] = {"0": "0", "1": "1"}
_lowerCamelCase , _lowerCamelCase : List[str] = "", ""
_lowerCamelCase : List[Any] = len(_lowerCAmelCase )
for i in range(len(_lowerCAmelCase ) ):
curr_string += data_bits[i]
if curr_string not in lexicon:
continue
_lowerCamelCase : Optional[Any] = lexicon[curr_string]
result += last_match_id
_lowerCamelCase : List[str] = last_match_id + "0"
if math.loga(_lowerCAmelCase ).is_integer():
_lowerCamelCase : Dict = {}
for curr_key in list(_lowerCAmelCase ):
_lowerCamelCase : Union[str, Any] = lexicon.pop(_lowerCAmelCase )
_lowerCamelCase : List[str] = new_lex
_lowerCamelCase : Union[str, Any] = last_match_id + "1"
index += 1
_lowerCamelCase : Union[str, Any] = ""
return result
def A_ ( _lowerCAmelCase : str , _lowerCAmelCase : str ):
"""simple docstring"""
_lowerCamelCase : str = 8
try:
with open(_lowerCAmelCase , "wb" ) as opened_file:
_lowerCamelCase : int = [
to_write[i : i + byte_length]
for i in range(0 , len(_lowerCAmelCase ) , _lowerCAmelCase )
]
if len(result_byte_array[-1] ) % byte_length == 0:
result_byte_array.append("10000000" )
else:
result_byte_array[-1] += "1" + "0" * (
byte_length - len(result_byte_array[-1] ) - 1
)
for elem in result_byte_array[:-1]:
opened_file.write(int(_lowerCAmelCase , 2 ).to_bytes(1 , byteorder="big" ) )
except OSError:
print("File not accessible" )
sys.exit()
def A_ ( _lowerCAmelCase : str ):
"""simple docstring"""
_lowerCamelCase : Optional[Any] = 0
for letter in data_bits:
if letter == "1":
break
counter += 1
_lowerCamelCase : Tuple = data_bits[counter:]
_lowerCamelCase : int = data_bits[counter + 1 :]
return data_bits
def A_ ( _lowerCAmelCase : str , _lowerCAmelCase : str ):
"""simple docstring"""
_lowerCamelCase : Dict = read_file_binary(_lowerCAmelCase )
_lowerCamelCase : List[str] = remove_prefix(_lowerCAmelCase )
_lowerCamelCase : List[Any] = decompress_data(_lowerCAmelCase )
write_file_binary(_lowerCAmelCase , _lowerCAmelCase )
if __name__ == "__main__":
compress(sys.argv[1], sys.argv[2]) | 44 |
import copy
from typing import TYPE_CHECKING, Any, Mapping, Optional, OrderedDict
from packaging import version
from ...configuration_utils import PretrainedConfig
from ...onnx import OnnxConfig
from ...utils import logging
from ..auto.configuration_auto import AutoConfig
if TYPE_CHECKING:
from ... import PreTrainedTokenizerBase, TensorType
snake_case__ : List[str] = logging.get_logger(__name__)
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
A_ = """vision-encoder-decoder"""
A_ = True
def __init__( self , **_UpperCAmelCase ) -> Dict:
super().__init__(**_UpperCAmelCase )
if "encoder" not in kwargs or "decoder" not in kwargs:
raise ValueError(
f"""A configuraton of type {self.model_type} cannot be instantiated because """
f"""not both `encoder` and `decoder` sub-configurations are passed, but only {kwargs}""" )
UpperCamelCase_ = kwargs.pop('encoder' )
UpperCamelCase_ = encoder_config.pop('model_type' )
UpperCamelCase_ = kwargs.pop('decoder' )
UpperCamelCase_ = decoder_config.pop('model_type' )
UpperCamelCase_ = AutoConfig.for_model(_UpperCAmelCase , **_UpperCAmelCase )
UpperCamelCase_ = AutoConfig.for_model(_UpperCAmelCase , **_UpperCAmelCase )
UpperCamelCase_ = True
@classmethod
def _UpperCAmelCase ( cls , _UpperCAmelCase , _UpperCAmelCase , **_UpperCAmelCase ) -> PretrainedConfig:
logger.info('Setting `config.is_decoder=True` and `config.add_cross_attention=True` for decoder_config' )
UpperCamelCase_ = True
UpperCamelCase_ = True
return cls(encoder=encoder_config.to_dict() , decoder=decoder_config.to_dict() , **_UpperCAmelCase )
def _UpperCAmelCase ( self ) -> int:
UpperCamelCase_ = copy.deepcopy(self.__dict__ )
UpperCamelCase_ = self.encoder.to_dict()
UpperCamelCase_ = self.decoder.to_dict()
UpperCamelCase_ = self.__class__.model_type
return output
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
A_ = version.parse("""1.11""" )
@property
def _UpperCAmelCase ( self ) -> Mapping[str, Mapping[int, str]]:
return OrderedDict(
[
('pixel_values', {0: 'batch', 1: 'num_channels', 2: 'height', 3: 'width'}),
] )
@property
def _UpperCAmelCase ( self ) -> float:
return 1e-4
@property
def _UpperCAmelCase ( self ) -> Mapping[str, Mapping[int, str]]:
return OrderedDict({'last_hidden_state': {0: 'batch', 1: 'encoder_sequence'}} )
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
@property
def _UpperCAmelCase ( self ) -> Mapping[str, Mapping[int, str]]:
UpperCamelCase_ = OrderedDict()
UpperCamelCase_ = {0: 'batch', 1: 'past_decoder_sequence + sequence'}
UpperCamelCase_ = {0: 'batch', 1: 'past_decoder_sequence + sequence'}
UpperCamelCase_ = {0: 'batch', 1: 'encoder_sequence'}
return common_inputs
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase = -1 , _UpperCAmelCase = -1 , _UpperCAmelCase = False , _UpperCAmelCase = None , ) -> Mapping[str, Any]:
import torch
UpperCamelCase_ = OrderedDict()
UpperCamelCase_ = super().generate_dummy_inputs(
_UpperCAmelCase , batch_size=_UpperCAmelCase , seq_length=_UpperCAmelCase , is_pair=_UpperCAmelCase , framework=_UpperCAmelCase )
UpperCamelCase_ , UpperCamelCase_ = dummy_input['input_ids'].shape
UpperCamelCase_ = (batch, encoder_sequence, self._config.encoder_hidden_size)
UpperCamelCase_ = dummy_input.pop('input_ids' )
UpperCamelCase_ = dummy_input.pop('attention_mask' )
UpperCamelCase_ = torch.zeros(_UpperCAmelCase )
return common_inputs
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
@property
def _UpperCAmelCase ( self ) -> None:
pass
def _UpperCAmelCase ( self , _UpperCAmelCase ) -> OnnxConfig:
return VisionEncoderDecoderEncoderOnnxConfig(_UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase = "default" ) -> OnnxConfig:
UpperCamelCase_ = encoder_config.hidden_size
return VisionEncoderDecoderDecoderOnnxConfig(_UpperCAmelCase , _UpperCAmelCase )
| 23 | 0 |
import unittest
from transformers import CamembertTokenizer, CamembertTokenizerFast
from transformers.testing_utils import get_tests_dir, require_sentencepiece, require_tokenizers, slow
from transformers.utils import is_torch_available
from ...test_tokenization_common import TokenizerTesterMixin
UpperCamelCase = get_tests_dir("fixtures/test_sentencepiece.model")
UpperCamelCase = get_tests_dir("fixtures/test_sentencepiece_bpe.model")
UpperCamelCase = "pt" if is_torch_available() else "tf"
@require_sentencepiece
@require_tokenizers
class lowerCAmelCase_ ( lowercase , unittest.TestCase ):
"""simple docstring"""
_snake_case : Dict = CamembertTokenizer
_snake_case : Optional[int] = CamembertTokenizerFast
_snake_case : Tuple = True
_snake_case : Dict = True
def __a ( self :Optional[int] ):
super().setUp()
# We have a SentencePiece fixture for testing
UpperCamelCase__ :Union[str, Any] = CamembertTokenizer(lowerCamelCase__ )
tokenizer.save_pretrained(self.tmpdirname )
def __a ( self :List[str] ):
UpperCamelCase__ :List[str] = """<pad>"""
UpperCamelCase__ :Dict = 1
self.assertEqual(self.get_tokenizer()._convert_token_to_id(lowerCamelCase__ ) , lowerCamelCase__ )
self.assertEqual(self.get_tokenizer()._convert_id_to_token(lowerCamelCase__ ) , lowerCamelCase__ )
def __a ( self :List[Any] ):
UpperCamelCase__ :Dict = list(self.get_tokenizer().get_vocab().keys() )
self.assertEqual(vocab_keys[0] , """<s>NOTUSED""" )
self.assertEqual(vocab_keys[1] , """<pad>""" )
self.assertEqual(vocab_keys[-1] , """<mask>""" )
self.assertEqual(len(lowerCamelCase__ ) , 10_04 )
def __a ( self :Optional[Any] ):
self.assertEqual(self.get_tokenizer().vocab_size , 10_05 )
def __a ( self :Dict ):
UpperCamelCase__ :List[str] = CamembertTokenizer(lowerCamelCase__ )
tokenizer.save_pretrained(self.tmpdirname )
UpperCamelCase__ :Union[str, Any] = CamembertTokenizerFast.from_pretrained(self.tmpdirname )
UpperCamelCase__ :Any = """I was born in 92000, and this is falsé."""
UpperCamelCase__ :List[Any] = tokenizer.encode(lowerCamelCase__ )
UpperCamelCase__ :Dict = rust_tokenizer.encode(lowerCamelCase__ )
self.assertListEqual(lowerCamelCase__ , lowerCamelCase__ )
UpperCamelCase__ :str = tokenizer.encode(lowerCamelCase__ , add_special_tokens=lowerCamelCase__ )
UpperCamelCase__ :Union[str, Any] = rust_tokenizer.encode(lowerCamelCase__ , add_special_tokens=lowerCamelCase__ )
self.assertListEqual(lowerCamelCase__ , lowerCamelCase__ )
# <unk> tokens are not the same for `rust` than for `slow`.
# Because spm gives back raw token instead of `unk` in EncodeAsPieces
# tokens = tokenizer.tokenize(sequence)
UpperCamelCase__ :Any = tokenizer.convert_ids_to_tokens(lowerCamelCase__ )
UpperCamelCase__ :str = rust_tokenizer.tokenize(lowerCamelCase__ )
self.assertListEqual(lowerCamelCase__ , lowerCamelCase__ )
def __a ( self :Union[str, Any] ):
if not self.test_rust_tokenizer:
return
UpperCamelCase__ :Dict = self.get_tokenizer()
UpperCamelCase__ :str = self.get_rust_tokenizer()
UpperCamelCase__ :str = """I was born in 92000, and this is falsé."""
UpperCamelCase__ :Optional[Any] = tokenizer.tokenize(lowerCamelCase__ )
UpperCamelCase__ :List[Any] = rust_tokenizer.tokenize(lowerCamelCase__ )
self.assertListEqual(lowerCamelCase__ , lowerCamelCase__ )
UpperCamelCase__ :str = tokenizer.encode(lowerCamelCase__ , add_special_tokens=lowerCamelCase__ )
UpperCamelCase__ :Union[str, Any] = rust_tokenizer.encode(lowerCamelCase__ , add_special_tokens=lowerCamelCase__ )
self.assertListEqual(lowerCamelCase__ , lowerCamelCase__ )
UpperCamelCase__ :Tuple = self.get_rust_tokenizer()
UpperCamelCase__ :Union[str, Any] = tokenizer.encode(lowerCamelCase__ )
UpperCamelCase__ :Optional[Any] = rust_tokenizer.encode(lowerCamelCase__ )
self.assertListEqual(lowerCamelCase__ , lowerCamelCase__ )
@slow
def __a ( self :List[str] ):
# fmt: off
UpperCamelCase__ :Any = {"""input_ids""": [[5, 54, 71_96, 2_97, 30, 23, 7_76, 18, 11, 32_15, 37_05, 82_52, 22, 31_64, 11_81, 21_16, 29, 16, 8_13, 25, 7_91, 33_14, 20, 34_46, 38, 2_75_75, 1_20, 6, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1], [5, 4_68, 17, 11, 90_88, 20, 15_17, 8, 2_28_04, 1_88_18, 10, 38, 6_29, 6_07, 6_07, 1_42, 19, 71_96, 8_67, 56, 1_03_26, 24, 22_67, 20, 4_16, 50_72, 1_56_12, 2_33, 7_34, 7, 23_99, 27, 16, 30_15, 16_49, 7, 24, 20, 43_38, 23_99, 27, 13, 34_00, 14, 13, 61_89, 8, 9_30, 9, 6]], """attention_mask""": [[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0], [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]]} # noqa: E501
# fmt: on
# camembert is a french model. So we also use french texts.
UpperCamelCase__ :Tuple = [
"""Le transformeur est un modèle d'apprentissage profond introduit en 2017, """
"""utilisé principalement dans le domaine du traitement automatique des langues (TAL).""",
"""À l'instar des réseaux de neurones récurrents (RNN), les transformeurs sont conçus """
"""pour gérer des données séquentielles, telles que le langage naturel, pour des tâches """
"""telles que la traduction et la synthèse de texte.""",
]
self.tokenizer_integration_test_util(
expected_encoding=lowerCamelCase__ , model_name="""camembert-base""" , revision="""3a0641d9a1aeb7e848a74299e7e4c4bca216b4cf""" , sequences=lowerCamelCase__ , ) | 45 |
import argparse
import torch
from transformers import MobileBertConfig, MobileBertForPreTraining, load_tf_weights_in_mobilebert
from transformers.utils import logging
logging.set_verbosity_info()
def _snake_case (__lowercase , __lowercase , __lowercase):
# Initialise PyTorch model
UpperCamelCase_ = MobileBertConfig.from_json_file(__lowercase)
print(f"""Building PyTorch model from configuration: {config}""")
UpperCamelCase_ = MobileBertForPreTraining(__lowercase)
# Load weights from tf checkpoint
UpperCamelCase_ = load_tf_weights_in_mobilebert(__lowercase , __lowercase , __lowercase)
# Save pytorch-model
print(f"""Save PyTorch model to {pytorch_dump_path}""")
torch.save(model.state_dict() , __lowercase)
if __name__ == "__main__":
snake_case__ : int = argparse.ArgumentParser()
# Required parameters
parser.add_argument(
"""--tf_checkpoint_path""", default=None, type=str, required=True, help="""Path to the TensorFlow checkpoint path."""
)
parser.add_argument(
"""--mobilebert_config_file""",
default=None,
type=str,
required=True,
help=(
"""The config json file corresponding to the pre-trained MobileBERT model. \n"""
"""This specifies the model architecture."""
),
)
parser.add_argument(
"""--pytorch_dump_path""", default=None, type=str, required=True, help="""Path to the output PyTorch model."""
)
snake_case__ : Optional[Any] = parser.parse_args()
convert_tf_checkpoint_to_pytorch(args.tf_checkpoint_path, args.mobilebert_config_file, args.pytorch_dump_path)
| 23 | 0 |
"""simple docstring"""
import importlib
import json
import os
import sys
import tempfile
import unittest
from pathlib import Path
import transformers
import transformers.models.auto
from transformers.models.auto.configuration_auto import CONFIG_MAPPING, AutoConfig
from transformers.models.bert.configuration_bert import BertConfig
from transformers.models.roberta.configuration_roberta import RobertaConfig
from transformers.testing_utils import DUMMY_UNKNOWN_IDENTIFIER, get_tests_dir
sys.path.append(str(Path(__file__).parent.parent.parent.parent / '''utils'''))
from test_module.custom_configuration import CustomConfig # noqa E402
_lowerCAmelCase : List[str] = get_tests_dir('''fixtures/dummy-config.json''')
class A_ ( unittest.TestCase ):
def _lowercase ( self: int ):
'''simple docstring'''
_lowerCamelCase : List[Any] = 0
def _lowercase ( self: Dict ):
'''simple docstring'''
self.assertIsNotNone(transformers.models.auto.__spec__ )
self.assertIsNotNone(importlib.util.find_spec("transformers.models.auto" ) )
def _lowercase ( self: List[Any] ):
'''simple docstring'''
_lowerCamelCase : Dict = AutoConfig.from_pretrained("bert-base-uncased" )
self.assertIsInstance(__lowerCAmelCase ,__lowerCAmelCase )
def _lowercase ( self: Dict ):
'''simple docstring'''
_lowerCamelCase : Optional[Any] = AutoConfig.from_pretrained(__lowerCAmelCase )
self.assertIsInstance(__lowerCAmelCase ,__lowerCAmelCase )
def _lowercase ( self: Any ):
'''simple docstring'''
_lowerCamelCase : List[str] = AutoConfig.from_pretrained(__lowerCAmelCase )
self.assertIsInstance(__lowerCAmelCase ,__lowerCAmelCase )
def _lowercase ( self: List[str] ):
'''simple docstring'''
_lowerCamelCase : Union[str, Any] = AutoConfig.for_model("roberta" )
self.assertIsInstance(__lowerCAmelCase ,__lowerCAmelCase )
def _lowercase ( self: List[Any] ):
'''simple docstring'''
with tempfile.TemporaryDirectory() as tmp_dir:
# This model name contains bert and roberta, but roberta ends up being picked.
_lowerCamelCase : List[Any] = os.path.join(__lowerCAmelCase ,"fake-roberta" )
os.makedirs(__lowerCAmelCase ,exist_ok=__lowerCAmelCase )
with open(os.path.join(__lowerCAmelCase ,"config.json" ) ,"w" ) as f:
f.write(json.dumps({} ) )
_lowerCamelCase : List[Any] = AutoConfig.from_pretrained(__lowerCAmelCase )
self.assertEqual(type(__lowerCAmelCase ) ,__lowerCAmelCase )
def _lowercase ( self: Dict ):
'''simple docstring'''
try:
AutoConfig.register("custom" ,__lowerCAmelCase )
# Wrong model type will raise an error
with self.assertRaises(__lowerCAmelCase ):
AutoConfig.register("model" ,__lowerCAmelCase )
# Trying to register something existing in the Transformers library will raise an error
with self.assertRaises(__lowerCAmelCase ):
AutoConfig.register("bert" ,__lowerCAmelCase )
# Now that the config is registered, it can be used as any other config with the auto-API
_lowerCamelCase : Any = CustomConfig()
with tempfile.TemporaryDirectory() as tmp_dir:
config.save_pretrained(__lowerCAmelCase )
_lowerCamelCase : List[str] = AutoConfig.from_pretrained(__lowerCAmelCase )
self.assertIsInstance(__lowerCAmelCase ,__lowerCAmelCase )
finally:
if "custom" in CONFIG_MAPPING._extra_content:
del CONFIG_MAPPING._extra_content["custom"]
def _lowercase ( self: Dict ):
'''simple docstring'''
with self.assertRaisesRegex(
__lowerCAmelCase ,"bert-base is not a local folder and is not a valid model identifier" ):
_lowerCamelCase : Union[str, Any] = AutoConfig.from_pretrained("bert-base" )
def _lowercase ( self: Dict ):
'''simple docstring'''
with self.assertRaisesRegex(
__lowerCAmelCase ,r"aaaaaa is not a valid git identifier \(branch name, tag name or commit id\)" ):
_lowerCamelCase : str = AutoConfig.from_pretrained(__lowerCAmelCase ,revision="aaaaaa" )
def _lowercase ( self: Tuple ):
'''simple docstring'''
with self.assertRaisesRegex(
__lowerCAmelCase ,"hf-internal-testing/no-config-test-repo does not appear to have a file named config.json." ,):
_lowerCamelCase : List[str] = AutoConfig.from_pretrained("hf-internal-testing/no-config-test-repo" )
def _lowercase ( self: List[Any] ):
'''simple docstring'''
with self.assertRaises(__lowerCAmelCase ):
_lowerCamelCase : List[Any] = AutoConfig.from_pretrained("hf-internal-testing/test_dynamic_model" )
# If remote code is disabled, we can't load this config.
with self.assertRaises(__lowerCAmelCase ):
_lowerCamelCase : Any = AutoConfig.from_pretrained("hf-internal-testing/test_dynamic_model" ,trust_remote_code=__lowerCAmelCase )
_lowerCamelCase : Optional[Any] = AutoConfig.from_pretrained("hf-internal-testing/test_dynamic_model" ,trust_remote_code=__lowerCAmelCase )
self.assertEqual(config.__class__.__name__ ,"NewModelConfig" )
# Test config can be reloaded.
with tempfile.TemporaryDirectory() as tmp_dir:
config.save_pretrained(__lowerCAmelCase )
_lowerCamelCase : str = AutoConfig.from_pretrained(__lowerCAmelCase ,trust_remote_code=__lowerCAmelCase )
self.assertEqual(reloaded_config.__class__.__name__ ,"NewModelConfig" )
def _lowercase ( self: Dict ):
'''simple docstring'''
class A_ ( _a ):
lowerCAmelCase__ = 'new-model'
try:
AutoConfig.register("new-model" ,__lowerCAmelCase )
# If remote code is not set, the default is to use local
_lowerCamelCase : List[Any] = AutoConfig.from_pretrained("hf-internal-testing/test_dynamic_model" )
self.assertEqual(config.__class__.__name__ ,"NewModelConfigLocal" )
# If remote code is disabled, we load the local one.
_lowerCamelCase : List[Any] = AutoConfig.from_pretrained("hf-internal-testing/test_dynamic_model" ,trust_remote_code=__lowerCAmelCase )
self.assertEqual(config.__class__.__name__ ,"NewModelConfigLocal" )
# If remote is enabled, we load from the Hub
_lowerCamelCase : List[Any] = AutoConfig.from_pretrained("hf-internal-testing/test_dynamic_model" ,trust_remote_code=__lowerCAmelCase )
self.assertEqual(config.__class__.__name__ ,"NewModelConfig" )
finally:
if "new-model" in CONFIG_MAPPING._extra_content:
del CONFIG_MAPPING._extra_content["new-model"] | 46 |
import gc
import unittest
from transformers import MODEL_FOR_MASKED_LM_MAPPING, TF_MODEL_FOR_MASKED_LM_MAPPING, FillMaskPipeline, pipeline
from transformers.pipelines import PipelineException
from transformers.testing_utils import (
is_pipeline_test,
is_torch_available,
nested_simplify,
require_tf,
require_torch,
require_torch_gpu,
slow,
)
from .test_pipelines_common import ANY
@is_pipeline_test
class _a ( unittest.TestCase ):
"""simple docstring"""
A_ = MODEL_FOR_MASKED_LM_MAPPING
A_ = TF_MODEL_FOR_MASKED_LM_MAPPING
def _UpperCAmelCase ( self ) -> List[str]:
super().tearDown()
# clean-up as much as possible GPU memory occupied by PyTorch
gc.collect()
if is_torch_available():
import torch
torch.cuda.empty_cache()
@require_tf
def _UpperCAmelCase ( self ) -> str:
UpperCamelCase_ = pipeline(task='fill-mask' , model='sshleifer/tiny-distilroberta-base' , top_k=2 , framework='tf' )
UpperCamelCase_ = unmasker('My name is <mask>' )
self.assertEqual(
nested_simplify(_UpperCAmelCase , decimals=6 ) , [
{'sequence': 'My name is grouped', 'score': 2.1e-05, 'token': 38015, 'token_str': ' grouped'},
{'sequence': 'My name is accuser', 'score': 2.1e-05, 'token': 25506, 'token_str': ' accuser'},
] , )
UpperCamelCase_ = unmasker('The largest city in France is <mask>' )
self.assertEqual(
nested_simplify(_UpperCAmelCase , decimals=6 ) , [
{
'sequence': 'The largest city in France is grouped',
'score': 2.1e-05,
'token': 38015,
'token_str': ' grouped',
},
{
'sequence': 'The largest city in France is accuser',
'score': 2.1e-05,
'token': 25506,
'token_str': ' accuser',
},
] , )
UpperCamelCase_ = unmasker('My name is <mask>' , targets=[' Patrick', ' Clara', ' Teven'] , top_k=3 )
self.assertEqual(
nested_simplify(_UpperCAmelCase , decimals=6 ) , [
{'sequence': 'My name is Clara', 'score': 2e-05, 'token': 13606, 'token_str': ' Clara'},
{'sequence': 'My name is Patrick', 'score': 2e-05, 'token': 3499, 'token_str': ' Patrick'},
{'sequence': 'My name is Te', 'score': 1.9e-05, 'token': 2941, 'token_str': ' Te'},
] , )
@require_torch
def _UpperCAmelCase ( self ) -> Union[str, Any]:
UpperCamelCase_ = pipeline(task='fill-mask' , model='sshleifer/tiny-distilroberta-base' , top_k=2 , framework='pt' )
UpperCamelCase_ = unmasker('My name is <mask>' )
self.assertEqual(
nested_simplify(_UpperCAmelCase , decimals=6 ) , [
{'sequence': 'My name is Maul', 'score': 2.2e-05, 'token': 35676, 'token_str': ' Maul'},
{'sequence': 'My name isELS', 'score': 2.2e-05, 'token': 16416, 'token_str': 'ELS'},
] , )
UpperCamelCase_ = unmasker('The largest city in France is <mask>' )
self.assertEqual(
nested_simplify(_UpperCAmelCase , decimals=6 ) , [
{
'sequence': 'The largest city in France is Maul',
'score': 2.2e-05,
'token': 35676,
'token_str': ' Maul',
},
{'sequence': 'The largest city in France isELS', 'score': 2.2e-05, 'token': 16416, 'token_str': 'ELS'},
] , )
UpperCamelCase_ = unmasker('My name is <mask>' , targets=[' Patrick', ' Clara', ' Teven'] , top_k=3 )
self.assertEqual(
nested_simplify(_UpperCAmelCase , decimals=6 ) , [
{'sequence': 'My name is Patrick', 'score': 2.1e-05, 'token': 3499, 'token_str': ' Patrick'},
{'sequence': 'My name is Te', 'score': 2e-05, 'token': 2941, 'token_str': ' Te'},
{'sequence': 'My name is Clara', 'score': 2e-05, 'token': 13606, 'token_str': ' Clara'},
] , )
UpperCamelCase_ = unmasker('My name is <mask> <mask>' , top_k=2 )
self.assertEqual(
nested_simplify(_UpperCAmelCase , decimals=6 ) , [
[
{
'score': 2.2e-05,
'token': 35676,
'token_str': ' Maul',
'sequence': '<s>My name is Maul<mask></s>',
},
{'score': 2.2e-05, 'token': 16416, 'token_str': 'ELS', 'sequence': '<s>My name isELS<mask></s>'},
],
[
{
'score': 2.2e-05,
'token': 35676,
'token_str': ' Maul',
'sequence': '<s>My name is<mask> Maul</s>',
},
{'score': 2.2e-05, 'token': 16416, 'token_str': 'ELS', 'sequence': '<s>My name is<mask>ELS</s>'},
],
] , )
@require_torch_gpu
def _UpperCAmelCase ( self ) -> Optional[Any]:
UpperCamelCase_ = pipeline('fill-mask' , model='hf-internal-testing/tiny-random-distilbert' , device=0 , framework='pt' )
# convert model to fp16
pipe.model.half()
UpperCamelCase_ = pipe('Paris is the [MASK] of France.' )
# We actually don't care about the result, we just want to make sure
# it works, meaning the float16 tensor got casted back to float32
# for postprocessing.
self.assertIsInstance(_UpperCAmelCase , _UpperCAmelCase )
@slow
@require_torch
def _UpperCAmelCase ( self ) -> List[Any]:
UpperCamelCase_ = pipeline(task='fill-mask' , model='distilroberta-base' , top_k=2 , framework='pt' )
self.run_large_test(_UpperCAmelCase )
@slow
@require_tf
def _UpperCAmelCase ( self ) -> Union[str, Any]:
UpperCamelCase_ = pipeline(task='fill-mask' , model='distilroberta-base' , top_k=2 , framework='tf' )
self.run_large_test(_UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase ) -> Tuple:
UpperCamelCase_ = unmasker('My name is <mask>' )
self.assertEqual(
nested_simplify(_UpperCAmelCase ) , [
{'sequence': 'My name is John', 'score': 0.0_0_8, 'token': 610, 'token_str': ' John'},
{'sequence': 'My name is Chris', 'score': 0.0_0_7, 'token': 1573, 'token_str': ' Chris'},
] , )
UpperCamelCase_ = unmasker('The largest city in France is <mask>' )
self.assertEqual(
nested_simplify(_UpperCAmelCase ) , [
{
'sequence': 'The largest city in France is Paris',
'score': 0.2_5_1,
'token': 2201,
'token_str': ' Paris',
},
{
'sequence': 'The largest city in France is Lyon',
'score': 0.2_1_4,
'token': 12790,
'token_str': ' Lyon',
},
] , )
UpperCamelCase_ = unmasker('My name is <mask>' , targets=[' Patrick', ' Clara', ' Teven'] , top_k=3 )
self.assertEqual(
nested_simplify(_UpperCAmelCase ) , [
{'sequence': 'My name is Patrick', 'score': 0.0_0_5, 'token': 3499, 'token_str': ' Patrick'},
{'sequence': 'My name is Clara', 'score': 0.0_0_0, 'token': 13606, 'token_str': ' Clara'},
{'sequence': 'My name is Te', 'score': 0.0_0_0, 'token': 2941, 'token_str': ' Te'},
] , )
@require_torch
def _UpperCAmelCase ( self ) -> List[Any]:
UpperCamelCase_ = pipeline(task='fill-mask' , model='sshleifer/tiny-distilroberta-base' , framework='pt' )
UpperCamelCase_ = None
UpperCamelCase_ = None
self.run_pipeline_test(_UpperCAmelCase , [] )
@require_tf
def _UpperCAmelCase ( self ) -> Optional[int]:
UpperCamelCase_ = pipeline(task='fill-mask' , model='sshleifer/tiny-distilroberta-base' , framework='tf' )
UpperCamelCase_ = None
UpperCamelCase_ = None
self.run_pipeline_test(_UpperCAmelCase , [] )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase ) -> Optional[Any]:
if tokenizer is None or tokenizer.mask_token_id is None:
self.skipTest('The provided tokenizer has no mask token, (probably reformer or wav2vec2)' )
UpperCamelCase_ = FillMaskPipeline(model=_UpperCAmelCase , tokenizer=_UpperCAmelCase )
UpperCamelCase_ = [
f"""This is another {tokenizer.mask_token} test""",
]
return fill_masker, examples
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> Union[str, Any]:
UpperCamelCase_ = fill_masker.tokenizer
UpperCamelCase_ = fill_masker.model
UpperCamelCase_ = fill_masker(
f"""This is a {tokenizer.mask_token}""" , )
self.assertEqual(
_UpperCAmelCase , [
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
] , )
UpperCamelCase_ = fill_masker([f"""This is a {tokenizer.mask_token}"""] )
self.assertEqual(
_UpperCAmelCase , [
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
] , )
UpperCamelCase_ = fill_masker([f"""This is a {tokenizer.mask_token}""", f"""Another {tokenizer.mask_token} great test."""] )
self.assertEqual(
_UpperCAmelCase , [
[
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
],
[
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
],
] , )
with self.assertRaises(_UpperCAmelCase ):
fill_masker([None] )
# No mask_token is not supported
with self.assertRaises(_UpperCAmelCase ):
fill_masker('This is' )
self.run_test_top_k(_UpperCAmelCase , _UpperCAmelCase )
self.run_test_targets(_UpperCAmelCase , _UpperCAmelCase )
self.run_test_top_k_targets(_UpperCAmelCase , _UpperCAmelCase )
self.fill_mask_with_duplicate_targets_and_top_k(_UpperCAmelCase , _UpperCAmelCase )
self.fill_mask_with_multiple_masks(_UpperCAmelCase , _UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> Optional[Any]:
UpperCamelCase_ = tokenizer.get_vocab()
UpperCamelCase_ = sorted(vocab.keys() )[:2]
# Pipeline argument
UpperCamelCase_ = FillMaskPipeline(model=_UpperCAmelCase , tokenizer=_UpperCAmelCase , targets=_UpperCAmelCase )
UpperCamelCase_ = fill_masker(f"""This is a {tokenizer.mask_token}""" )
self.assertEqual(
_UpperCAmelCase , [
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
] , )
UpperCamelCase_ = {vocab[el] for el in targets}
self.assertEqual({el['token'] for el in outputs} , _UpperCAmelCase )
UpperCamelCase_ = [tokenizer.decode([x] ) for x in target_ids]
self.assertEqual({el['token_str'] for el in outputs} , set(_UpperCAmelCase ) )
# Call argument
UpperCamelCase_ = FillMaskPipeline(model=_UpperCAmelCase , tokenizer=_UpperCAmelCase )
UpperCamelCase_ = fill_masker(f"""This is a {tokenizer.mask_token}""" , targets=_UpperCAmelCase )
self.assertEqual(
_UpperCAmelCase , [
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
] , )
UpperCamelCase_ = {vocab[el] for el in targets}
self.assertEqual({el['token'] for el in outputs} , _UpperCAmelCase )
UpperCamelCase_ = [tokenizer.decode([x] ) for x in target_ids]
self.assertEqual({el['token_str'] for el in outputs} , set(_UpperCAmelCase ) )
# Score equivalence
UpperCamelCase_ = fill_masker(f"""This is a {tokenizer.mask_token}""" , targets=_UpperCAmelCase )
UpperCamelCase_ = [top_mask['token_str'] for top_mask in outputs]
UpperCamelCase_ = [top_mask['score'] for top_mask in outputs]
# For some BPE tokenizers, `</w>` is removed during decoding, so `token_str` won't be the same as in `targets`.
if set(_UpperCAmelCase ) == set(_UpperCAmelCase ):
UpperCamelCase_ = fill_masker(f"""This is a {tokenizer.mask_token}""" , targets=_UpperCAmelCase )
UpperCamelCase_ = [top_mask['score'] for top_mask in unmasked_targets]
self.assertEqual(nested_simplify(_UpperCAmelCase ) , nested_simplify(_UpperCAmelCase ) )
# Raises with invalid
with self.assertRaises(_UpperCAmelCase ):
UpperCamelCase_ = fill_masker(f"""This is a {tokenizer.mask_token}""" , targets=[] )
# For some tokenizers, `""` is actually in the vocabulary and the expected error won't raised
if "" not in tokenizer.get_vocab():
with self.assertRaises(_UpperCAmelCase ):
UpperCamelCase_ = fill_masker(f"""This is a {tokenizer.mask_token}""" , targets=[''] )
with self.assertRaises(_UpperCAmelCase ):
UpperCamelCase_ = fill_masker(f"""This is a {tokenizer.mask_token}""" , targets='' )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> Tuple:
UpperCamelCase_ = FillMaskPipeline(model=_UpperCAmelCase , tokenizer=_UpperCAmelCase , top_k=2 )
UpperCamelCase_ = fill_masker(f"""This is a {tokenizer.mask_token}""" )
self.assertEqual(
_UpperCAmelCase , [
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
] , )
UpperCamelCase_ = FillMaskPipeline(model=_UpperCAmelCase , tokenizer=_UpperCAmelCase )
UpperCamelCase_ = fill_masker(f"""This is a {tokenizer.mask_token}""" , top_k=2 )
self.assertEqual(
_UpperCAmelCase , [
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
] , )
self.assertEqual(nested_simplify(_UpperCAmelCase ) , nested_simplify(_UpperCAmelCase ) )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> List[Any]:
UpperCamelCase_ = tokenizer.get_vocab()
UpperCamelCase_ = FillMaskPipeline(model=_UpperCAmelCase , tokenizer=_UpperCAmelCase )
# top_k=2, ntargets=3
UpperCamelCase_ = sorted(vocab.keys() )[:3]
UpperCamelCase_ = fill_masker(f"""This is a {tokenizer.mask_token}""" , top_k=2 , targets=_UpperCAmelCase )
# If we use the most probably targets, and filter differently, we should still
# have the same results
UpperCamelCase_ = [el['token_str'] for el in sorted(_UpperCAmelCase , key=lambda _UpperCAmelCase : x["score"] , reverse=_UpperCAmelCase )]
# For some BPE tokenizers, `</w>` is removed during decoding, so `token_str` won't be the same as in `targets`.
if set(_UpperCAmelCase ).issubset(_UpperCAmelCase ):
UpperCamelCase_ = fill_masker(f"""This is a {tokenizer.mask_token}""" , top_k=3 , targets=_UpperCAmelCase )
# They should yield exactly the same result
self.assertEqual(nested_simplify(_UpperCAmelCase ) , nested_simplify(_UpperCAmelCase ) )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> Optional[int]:
UpperCamelCase_ = FillMaskPipeline(model=_UpperCAmelCase , tokenizer=_UpperCAmelCase )
UpperCamelCase_ = tokenizer.get_vocab()
# String duplicates + id duplicates
UpperCamelCase_ = sorted(vocab.keys() )[:3]
UpperCamelCase_ = [targets[0], targets[1], targets[0], targets[2], targets[1]]
UpperCamelCase_ = fill_masker(f"""My name is {tokenizer.mask_token}""" , targets=_UpperCAmelCase , top_k=10 )
# The target list contains duplicates, so we can't output more
# than them
self.assertEqual(len(_UpperCAmelCase ) , 3 )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> List[str]:
UpperCamelCase_ = FillMaskPipeline(model=_UpperCAmelCase , tokenizer=_UpperCAmelCase )
UpperCamelCase_ = fill_masker(
f"""This is a {tokenizer.mask_token} {tokenizer.mask_token} {tokenizer.mask_token}""" , top_k=2 )
self.assertEqual(
_UpperCAmelCase , [
[
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
],
[
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
],
[
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
{'sequence': ANY(_UpperCAmelCase ), 'score': ANY(_UpperCAmelCase ), 'token': ANY(_UpperCAmelCase ), 'token_str': ANY(_UpperCAmelCase )},
],
] , )
| 23 | 0 |
import argparse
import os
import evaluate
import torch
from datasets import load_dataset
from torch.optim import AdamW
from torch.utils.data import DataLoader
from transformers import AutoModelForSequenceClassification, AutoTokenizer, get_linear_schedule_with_warmup, set_seed
from accelerate import Accelerator, DistributedType
########################################################################
# This is a fully working simple example to use Accelerate
# and perform gradient accumulation
#
# This example trains a Bert base model on GLUE MRPC
# in any of the following settings (with the same script):
# - single CPU or single GPU
# - multi GPUS (using PyTorch distributed mode)
# - (multi) TPUs
# - fp16 (mixed-precision) or fp32 (normal precision)
#
# To run it in each of these various modes, follow the instructions
# in the readme for examples:
# https://github.com/huggingface/accelerate/tree/main/examples
#
########################################################################
SCREAMING_SNAKE_CASE__ = 16
SCREAMING_SNAKE_CASE__ = 32
def UpperCAmelCase__ ( lowerCamelCase_ : Accelerator , lowerCamelCase_ : int = 1_6 ):
__a : List[Any] = AutoTokenizer.from_pretrained('bert-base-cased' )
__a : int = load_dataset('glue' , 'mrpc' )
def tokenize_function(lowerCamelCase_ : int ):
# max_length=None => use the model max length (it's actually the default)
__a : Dict = tokenizer(examples['sentence1'] , examples['sentence2'] , truncation=lowerCamelCase_ , max_length=lowerCamelCase_ )
return outputs
# Apply the method we just defined to all the examples in all the splits of the dataset
# starting with the main process first:
with accelerator.main_process_first():
__a : str = datasets.map(
lowerCamelCase_ , batched=lowerCamelCase_ , remove_columns=['idx', 'sentence1', 'sentence2'] , )
# We also rename the 'label' column to 'labels' which is the expected name for labels by the models of the
# transformers library
__a : Union[str, Any] = tokenized_datasets.rename_column('label' , 'labels' )
def collate_fn(lowerCamelCase_ : Union[str, Any] ):
# On TPU it's best to pad everything to the same length or training will be very slow.
__a : Tuple = 1_2_8 if accelerator.distributed_type == DistributedType.TPU else None
# When using mixed precision we want round multiples of 8/16
if accelerator.mixed_precision == "fp8":
__a : Any = 1_6
elif accelerator.mixed_precision != "no":
__a : Optional[Any] = 8
else:
__a : str = None
return tokenizer.pad(
lowerCamelCase_ , padding='longest' , max_length=lowerCamelCase_ , pad_to_multiple_of=lowerCamelCase_ , return_tensors='pt' , )
# Instantiate dataloaders.
__a : Union[str, Any] = DataLoader(
tokenized_datasets['train'] , shuffle=lowerCamelCase_ , collate_fn=lowerCamelCase_ , batch_size=lowerCamelCase_ )
__a : Dict = DataLoader(
tokenized_datasets['validation'] , shuffle=lowerCamelCase_ , collate_fn=lowerCamelCase_ , batch_size=lowerCamelCase_ )
return train_dataloader, eval_dataloader
# For testing only
if os.environ.get('''TESTING_MOCKED_DATALOADERS''', None) == "1":
from accelerate.test_utils.training import mocked_dataloaders
SCREAMING_SNAKE_CASE__ = mocked_dataloaders # noqa: F811
def UpperCAmelCase__ ( lowerCamelCase_ : Tuple , lowerCamelCase_ : Dict ):
# For testing only
if os.environ.get('TESTING_MOCKED_DATALOADERS' , lowerCamelCase_ ) == "1":
__a : List[str] = 2
# New Code #
__a : Dict = int(args.gradient_accumulation_steps )
# Initialize accelerator
__a : Optional[Any] = Accelerator(
cpu=args.cpu , mixed_precision=args.mixed_precision , gradient_accumulation_steps=lowerCamelCase_ )
if accelerator.distributed_type == DistributedType.TPU and gradient_accumulation_steps > 1:
raise NotImplementedError(
'Gradient accumulation on TPUs is currently not supported. Pass `gradient_accumulation_steps=1`' )
# Sample hyper-parameters for learning rate, batch size, seed and a few other HPs
__a : Union[str, Any] = config['lr']
__a : List[Any] = int(config['num_epochs'] )
__a : Optional[int] = int(config['seed'] )
__a : Dict = int(config['batch_size'] )
__a : Optional[int] = evaluate.load('glue' , 'mrpc' )
set_seed(lowerCamelCase_ )
__a , __a : List[Any] = get_dataloaders(lowerCamelCase_ , lowerCamelCase_ )
# Instantiate the model (we build the model here so that the seed also control new weights initialization)
__a : int = AutoModelForSequenceClassification.from_pretrained('bert-base-cased' , return_dict=lowerCamelCase_ )
# We could avoid this line since the accelerator is set with `device_placement=True` (default value).
# Note that if you are placing tensors on devices manually, this line absolutely needs to be before the optimizer
# creation otherwise training will not work on TPU (`accelerate` will kindly throw an error to make us aware of that).
__a : Any = model.to(accelerator.device )
# Instantiate optimizer
__a : Dict = AdamW(params=model.parameters() , lr=lowerCamelCase_ )
# Instantiate scheduler
__a : List[str] = get_linear_schedule_with_warmup(
optimizer=lowerCamelCase_ , num_warmup_steps=1_0_0 , num_training_steps=(len(lowerCamelCase_ ) * num_epochs) , )
# Prepare everything
# There is no specific order to remember, we just need to unpack the objects in the same order we gave them to the
# prepare method.
__a , __a , __a , __a , __a : Dict = accelerator.prepare(
lowerCamelCase_ , lowerCamelCase_ , lowerCamelCase_ , lowerCamelCase_ , lowerCamelCase_ )
# Now we train the model
for epoch in range(lowerCamelCase_ ):
model.train()
for step, batch in enumerate(lowerCamelCase_ ):
# We could avoid this line since we set the accelerator with `device_placement=True`.
batch.to(accelerator.device )
# New code #
# We use the new `accumulate` context manager to perform gradient accumulation
# We also currently do not support TPUs nor advise it as bugs were found on the XLA side when running our tests.
with accelerator.accumulate(lowerCamelCase_ ):
__a : Any = model(**lowerCamelCase_ )
__a : Dict = output.loss
accelerator.backward(lowerCamelCase_ )
optimizer.step()
lr_scheduler.step()
optimizer.zero_grad()
model.eval()
for step, batch in enumerate(lowerCamelCase_ ):
# We could avoid this line since we set the accelerator with `device_placement=True`.
batch.to(accelerator.device )
with torch.no_grad():
__a : Any = model(**lowerCamelCase_ )
__a : Optional[Any] = outputs.logits.argmax(dim=-1 )
__a , __a : Dict = accelerator.gather_for_metrics((predictions, batch['labels']) )
metric.add_batch(
predictions=lowerCamelCase_ , references=lowerCamelCase_ , )
__a : Optional[int] = metric.compute()
# Use accelerator.print to print only on the main process.
accelerator.print(f'''epoch {epoch}:''' , lowerCamelCase_ )
def UpperCAmelCase__ ( ):
__a : Union[str, Any] = argparse.ArgumentParser(description='Simple example of training script.' )
parser.add_argument(
'--mixed_precision' , type=lowerCamelCase_ , default=lowerCamelCase_ , choices=['no', 'fp16', 'bf16', 'fp8'] , help='Whether to use mixed precision. Choose'
'between fp16 and bf16 (bfloat16). Bf16 requires PyTorch >= 1.10.'
'and an Nvidia Ampere GPU.' , )
# New Code #
parser.add_argument(
'--gradient_accumulation_steps' , type=lowerCamelCase_ , default=1 , help='The number of minibatches to be ran before gradients are accumulated.' , )
parser.add_argument('--cpu' , action='store_true' , help='If passed, will train on the CPU.' )
__a : Any = parser.parse_args()
__a : Optional[Any] = {'lr': 2e-5, 'num_epochs': 3, 'seed': 4_2, 'batch_size': 1_6}
training_function(lowerCamelCase_ , lowerCamelCase_ )
if __name__ == "__main__":
main()
| 47 |
import gc
import unittest
import numpy as np
import torch
from transformers import CLIPTextConfig, CLIPTextModel, CLIPTokenizer
from diffusers import (
AutoencoderKL,
DDIMScheduler,
StableDiffusionSAGPipeline,
UNetaDConditionModel,
)
from diffusers.utils import slow, torch_device
from diffusers.utils.testing_utils import enable_full_determinism, require_torch_gpu
from ..pipeline_params import TEXT_TO_IMAGE_BATCH_PARAMS, TEXT_TO_IMAGE_IMAGE_PARAMS, TEXT_TO_IMAGE_PARAMS
from ..test_pipelines_common import PipelineLatentTesterMixin, PipelineTesterMixin
enable_full_determinism()
class _a ( UpperCAmelCase__ , UpperCAmelCase__ , unittest.TestCase ):
"""simple docstring"""
A_ = StableDiffusionSAGPipeline
A_ = TEXT_TO_IMAGE_PARAMS
A_ = TEXT_TO_IMAGE_BATCH_PARAMS
A_ = TEXT_TO_IMAGE_IMAGE_PARAMS
A_ = TEXT_TO_IMAGE_IMAGE_PARAMS
A_ = False
def _UpperCAmelCase ( self ) -> Optional[Any]:
torch.manual_seed(0 )
UpperCamelCase_ = UNetaDConditionModel(
block_out_channels=(32, 64) , layers_per_block=2 , sample_size=32 , in_channels=4 , out_channels=4 , down_block_types=('DownBlock2D', 'CrossAttnDownBlock2D') , up_block_types=('CrossAttnUpBlock2D', 'UpBlock2D') , cross_attention_dim=32 , )
UpperCamelCase_ = DDIMScheduler(
beta_start=0.0_0_0_8_5 , beta_end=0.0_1_2 , beta_schedule='scaled_linear' , clip_sample=_UpperCAmelCase , set_alpha_to_one=_UpperCAmelCase , )
torch.manual_seed(0 )
UpperCamelCase_ = AutoencoderKL(
block_out_channels=[32, 64] , in_channels=3 , out_channels=3 , down_block_types=['DownEncoderBlock2D', 'DownEncoderBlock2D'] , up_block_types=['UpDecoderBlock2D', 'UpDecoderBlock2D'] , latent_channels=4 , )
torch.manual_seed(0 )
UpperCamelCase_ = CLIPTextConfig(
bos_token_id=0 , eos_token_id=2 , hidden_size=32 , intermediate_size=37 , layer_norm_eps=1e-05 , num_attention_heads=4 , num_hidden_layers=5 , pad_token_id=1 , vocab_size=1000 , )
UpperCamelCase_ = CLIPTextModel(_UpperCAmelCase )
UpperCamelCase_ = CLIPTokenizer.from_pretrained('hf-internal-testing/tiny-random-clip' )
UpperCamelCase_ = {
'unet': unet,
'scheduler': scheduler,
'vae': vae,
'text_encoder': text_encoder,
'tokenizer': tokenizer,
'safety_checker': None,
'feature_extractor': None,
}
return components
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase=0 ) -> List[Any]:
if str(_UpperCAmelCase ).startswith('mps' ):
UpperCamelCase_ = torch.manual_seed(_UpperCAmelCase )
else:
UpperCamelCase_ = torch.Generator(device=_UpperCAmelCase ).manual_seed(_UpperCAmelCase )
UpperCamelCase_ = {
'prompt': '.',
'generator': generator,
'num_inference_steps': 2,
'guidance_scale': 1.0,
'sag_scale': 1.0,
'output_type': 'numpy',
}
return inputs
def _UpperCAmelCase ( self ) -> Tuple:
super().test_inference_batch_single_identical(expected_max_diff=3e-3 )
@slow
@require_torch_gpu
class _a ( unittest.TestCase ):
"""simple docstring"""
def _UpperCAmelCase ( self ) -> Tuple:
# clean up the VRAM after each test
super().tearDown()
gc.collect()
torch.cuda.empty_cache()
def _UpperCAmelCase ( self ) -> str:
UpperCamelCase_ = StableDiffusionSAGPipeline.from_pretrained('CompVis/stable-diffusion-v1-4' )
UpperCamelCase_ = sag_pipe.to(_UpperCAmelCase )
sag_pipe.set_progress_bar_config(disable=_UpperCAmelCase )
UpperCamelCase_ = '.'
UpperCamelCase_ = torch.manual_seed(0 )
UpperCamelCase_ = sag_pipe(
[prompt] , generator=_UpperCAmelCase , guidance_scale=7.5 , sag_scale=1.0 , num_inference_steps=20 , output_type='np' )
UpperCamelCase_ = output.images
UpperCamelCase_ = image[0, -3:, -3:, -1]
assert image.shape == (1, 512, 512, 3)
UpperCamelCase_ = np.array([0.1_5_6_8, 0.1_7_3_8, 0.1_6_9_5, 0.1_6_9_3, 0.1_5_0_7, 0.1_7_0_5, 0.1_5_4_7, 0.1_7_5_1, 0.1_9_4_9] )
assert np.abs(image_slice.flatten() - expected_slice ).max() < 5e-2
def _UpperCAmelCase ( self ) -> List[Any]:
UpperCamelCase_ = StableDiffusionSAGPipeline.from_pretrained('stabilityai/stable-diffusion-2-1-base' )
UpperCamelCase_ = sag_pipe.to(_UpperCAmelCase )
sag_pipe.set_progress_bar_config(disable=_UpperCAmelCase )
UpperCamelCase_ = '.'
UpperCamelCase_ = torch.manual_seed(0 )
UpperCamelCase_ = sag_pipe(
[prompt] , generator=_UpperCAmelCase , guidance_scale=7.5 , sag_scale=1.0 , num_inference_steps=20 , output_type='np' )
UpperCamelCase_ = output.images
UpperCamelCase_ = image[0, -3:, -3:, -1]
assert image.shape == (1, 512, 512, 3)
UpperCamelCase_ = np.array([0.3_4_5_9, 0.2_8_7_6, 0.2_5_3_7, 0.3_0_0_2, 0.2_6_7_1, 0.2_1_6_0, 0.3_0_2_6, 0.2_2_6_2, 0.2_3_7_1] )
assert np.abs(image_slice.flatten() - expected_slice ).max() < 5e-2
def _UpperCAmelCase ( self ) -> Dict:
UpperCamelCase_ = StableDiffusionSAGPipeline.from_pretrained('stabilityai/stable-diffusion-2-1-base' )
UpperCamelCase_ = sag_pipe.to(_UpperCAmelCase )
sag_pipe.set_progress_bar_config(disable=_UpperCAmelCase )
UpperCamelCase_ = '.'
UpperCamelCase_ = torch.manual_seed(0 )
UpperCamelCase_ = sag_pipe(
[prompt] , width=768 , height=512 , generator=_UpperCAmelCase , guidance_scale=7.5 , sag_scale=1.0 , num_inference_steps=20 , output_type='np' , )
UpperCamelCase_ = output.images
assert image.shape == (1, 512, 768, 3)
| 23 | 0 |
'''simple docstring'''
import warnings
from ...processing_utils import ProcessorMixin
from ...tokenization_utils_base import BatchEncoding
class A ( SCREAMING_SNAKE_CASE__ ):
snake_case__ :Tuple = ['image_processor', 'tokenizer']
snake_case__ :List[Any] = 'ChineseCLIPImageProcessor'
snake_case__ :Optional[int] = ('BertTokenizer', 'BertTokenizerFast')
def __init__( self : Dict , __magic_name__ : List[str]=None , __magic_name__ : List[Any]=None , **__magic_name__ : Optional[Any] ):
"""simple docstring"""
lowerCAmelCase__ = None
if "feature_extractor" in kwargs:
warnings.warn(
"The `feature_extractor` argument is deprecated and will be removed in v5, use `image_processor`"
" instead." , __magic_name__ , )
lowerCAmelCase__ = kwargs.pop("feature_extractor" )
lowerCAmelCase__ = image_processor if image_processor is not None else feature_extractor
if image_processor is None:
raise ValueError("You need to specify an `image_processor`." )
if tokenizer is None:
raise ValueError("You need to specify a `tokenizer`." )
super().__init__(__magic_name__ , __magic_name__ )
lowerCAmelCase__ = self.image_processor
def __call__( self : List[Any] , __magic_name__ : Tuple=None , __magic_name__ : Any=None , __magic_name__ : str=None , **__magic_name__ : List[str] ):
"""simple docstring"""
if text is None and images is None:
raise ValueError("You have to specify either text or images. Both cannot be none." )
if text is not None:
lowerCAmelCase__ = self.tokenizer(__magic_name__ , return_tensors=__magic_name__ , **__magic_name__ )
if images is not None:
lowerCAmelCase__ = self.image_processor(__magic_name__ , return_tensors=__magic_name__ , **__magic_name__ )
if text is not None and images is not None:
lowerCAmelCase__ = image_features.pixel_values
return encoding
elif text is not None:
return encoding
else:
return BatchEncoding(data=dict(**__magic_name__ ) , tensor_type=__magic_name__ )
def __SCREAMING_SNAKE_CASE ( self : Tuple , *__magic_name__ : Union[str, Any] , **__magic_name__ : List[str] ):
"""simple docstring"""
return self.tokenizer.batch_decode(*__magic_name__ , **__magic_name__ )
def __SCREAMING_SNAKE_CASE ( self : Any , *__magic_name__ : str , **__magic_name__ : Union[str, Any] ):
"""simple docstring"""
return self.tokenizer.decode(*__magic_name__ , **__magic_name__ )
@property
def __SCREAMING_SNAKE_CASE ( self : str ):
"""simple docstring"""
lowerCAmelCase__ = self.tokenizer.model_input_names
lowerCAmelCase__ = self.image_processor.model_input_names
return list(dict.fromkeys(tokenizer_input_names + image_processor_input_names ) )
@property
def __SCREAMING_SNAKE_CASE ( self : Tuple ):
"""simple docstring"""
warnings.warn(
"`feature_extractor_class` is deprecated and will be removed in v5. Use `image_processor_class` instead." , __magic_name__ , )
return self.image_processor_class
| 48 |
from __future__ import annotations
from sys import maxsize
from typing import Generic, TypeVar
snake_case__ : List[str] = TypeVar("""T""")
def _snake_case (__lowercase):
return (position - 1) // 2
def _snake_case (__lowercase):
return (2 * position) + 1
def _snake_case (__lowercase):
return (2 * position) + 2
class _a ( Generic[T] ):
"""simple docstring"""
def __init__( self ) -> None:
UpperCamelCase_ = []
UpperCamelCase_ = {}
UpperCamelCase_ = 0
def __len__( self ) -> int:
return self.elements
def __repr__( self ) -> str:
return str(self.heap )
def _UpperCAmelCase ( self ) -> bool:
# Check if the priority queue is empty
return self.elements == 0
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> None:
# Add an element with given priority to the queue
self.heap.append((elem, weight) )
UpperCamelCase_ = self.elements
self.elements += 1
self._bubble_up(_UpperCAmelCase )
def _UpperCAmelCase ( self ) -> T:
# Remove and return the element with lowest weight (highest priority)
if self.elements > 1:
self._swap_nodes(0 , self.elements - 1 )
UpperCamelCase_ , UpperCamelCase_ = self.heap.pop()
del self.position_map[elem]
self.elements -= 1
if self.elements > 0:
UpperCamelCase_ , UpperCamelCase_ = self.heap[0]
self._bubble_down(_UpperCAmelCase )
return elem
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> None:
# Update the weight of the given key
UpperCamelCase_ = self.position_map[elem]
UpperCamelCase_ = (elem, weight)
if position > 0:
UpperCamelCase_ = get_parent_position(_UpperCAmelCase )
UpperCamelCase_ , UpperCamelCase_ = self.heap[parent_position]
if parent_weight > weight:
self._bubble_up(_UpperCAmelCase )
else:
self._bubble_down(_UpperCAmelCase )
else:
self._bubble_down(_UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase ) -> None:
# Place a node at the proper position (upward movement) [to be used internally
# only]
UpperCamelCase_ = self.position_map[elem]
if curr_pos == 0:
return None
UpperCamelCase_ = get_parent_position(_UpperCAmelCase )
UpperCamelCase_ , UpperCamelCase_ = self.heap[curr_pos]
UpperCamelCase_ , UpperCamelCase_ = self.heap[parent_position]
if parent_weight > weight:
self._swap_nodes(_UpperCAmelCase , _UpperCAmelCase )
return self._bubble_up(_UpperCAmelCase )
return None
def _UpperCAmelCase ( self , _UpperCAmelCase ) -> None:
# Place a node at the proper position (downward movement) [to be used
# internally only]
UpperCamelCase_ = self.position_map[elem]
UpperCamelCase_ , UpperCamelCase_ = self.heap[curr_pos]
UpperCamelCase_ = get_child_left_position(_UpperCAmelCase )
UpperCamelCase_ = get_child_right_position(_UpperCAmelCase )
if child_left_position < self.elements and child_right_position < self.elements:
UpperCamelCase_ , UpperCamelCase_ = self.heap[child_left_position]
UpperCamelCase_ , UpperCamelCase_ = self.heap[child_right_position]
if child_right_weight < child_left_weight and child_right_weight < weight:
self._swap_nodes(_UpperCAmelCase , _UpperCAmelCase )
return self._bubble_down(_UpperCAmelCase )
if child_left_position < self.elements:
UpperCamelCase_ , UpperCamelCase_ = self.heap[child_left_position]
if child_left_weight < weight:
self._swap_nodes(_UpperCAmelCase , _UpperCAmelCase )
return self._bubble_down(_UpperCAmelCase )
else:
return None
if child_right_position < self.elements:
UpperCamelCase_ , UpperCamelCase_ = self.heap[child_right_position]
if child_right_weight < weight:
self._swap_nodes(_UpperCAmelCase , _UpperCAmelCase )
return self._bubble_down(_UpperCAmelCase )
return None
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> None:
# Swap the nodes at the given positions
UpperCamelCase_ = self.heap[nodea_pos][0]
UpperCamelCase_ = self.heap[nodea_pos][0]
UpperCamelCase_ , UpperCamelCase_ = (
self.heap[nodea_pos],
self.heap[nodea_pos],
)
UpperCamelCase_ = nodea_pos
UpperCamelCase_ = nodea_pos
class _a ( Generic[T] ):
"""simple docstring"""
def __init__( self ) -> None:
UpperCamelCase_ = {}
UpperCamelCase_ = 0
def __repr__( self ) -> str:
return str(self.connections )
def __len__( self ) -> int:
return self.nodes
def _UpperCAmelCase ( self , _UpperCAmelCase ) -> None:
# Add a node in the graph if it is not in the graph
if node not in self.connections:
UpperCamelCase_ = {}
self.nodes += 1
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase ) -> None:
# Add an edge between 2 nodes in the graph
self.add_node(_UpperCAmelCase )
self.add_node(_UpperCAmelCase )
UpperCamelCase_ = weight
UpperCamelCase_ = weight
def _snake_case (__lowercase , ):
UpperCamelCase_ = {node: maxsize for node in graph.connections}
UpperCamelCase_ = {node: None for node in graph.connections}
UpperCamelCase_ = MinPriorityQueue()
for node, weight in dist.items():
priority_queue.push(__lowercase , __lowercase)
if priority_queue.is_empty():
return dist, parent
# initialization
UpperCamelCase_ = priority_queue.extract_min()
UpperCamelCase_ = 0
for neighbour in graph.connections[node]:
if dist[neighbour] > dist[node] + graph.connections[node][neighbour]:
UpperCamelCase_ = dist[node] + graph.connections[node][neighbour]
priority_queue.update_key(__lowercase , dist[neighbour])
UpperCamelCase_ = node
# running prim's algorithm
while not priority_queue.is_empty():
UpperCamelCase_ = priority_queue.extract_min()
for neighbour in graph.connections[node]:
if dist[neighbour] > dist[node] + graph.connections[node][neighbour]:
UpperCamelCase_ = dist[node] + graph.connections[node][neighbour]
priority_queue.update_key(__lowercase , dist[neighbour])
UpperCamelCase_ = node
return dist, parent
| 23 | 0 |
"""simple docstring"""
import unittest
from transformers import is_vision_available
from transformers.pipelines import pipeline
from transformers.testing_utils import (
is_pipeline_test,
nested_simplify,
require_tf,
require_torch,
require_vision,
slow,
)
from .test_pipelines_common import ANY
if is_vision_available():
from PIL import Image
else:
class _UpperCAmelCase :
@staticmethod
def a ( *_lowercase : Optional[int] , **_lowercase : List[str] ):
pass
@is_pipeline_test
@require_vision
class _UpperCAmelCase ( unittest.TestCase ):
@require_torch
def a ( self : Dict ):
__UpperCAmelCase = pipeline(
model='''hf-internal-testing/tiny-random-clip-zero-shot-image-classification''' , )
__UpperCAmelCase = Image.open('''./tests/fixtures/tests_samples/COCO/000000039769.png''' )
__UpperCAmelCase = image_classifier(_lowercase , candidate_labels=['''a''', '''b''', '''c'''] )
# The floating scores are so close, we enter floating error approximation and the order is not guaranteed across
# python and torch versions.
self.assertIn(
nested_simplify(_lowercase ) , [
[{'''score''': 0.333, '''label''': '''a'''}, {'''score''': 0.333, '''label''': '''b'''}, {'''score''': 0.333, '''label''': '''c'''}],
[{'''score''': 0.333, '''label''': '''a'''}, {'''score''': 0.333, '''label''': '''c'''}, {'''score''': 0.333, '''label''': '''b'''}],
] , )
__UpperCAmelCase = image_classifier([image] * 5 , candidate_labels=['''A''', '''B''', '''C'''] , batch_size=2 )
self.assertEqual(
nested_simplify(_lowercase ) , [
[
{'''score''': 0.333, '''label''': ANY(_lowercase )},
{'''score''': 0.333, '''label''': ANY(_lowercase )},
{'''score''': 0.333, '''label''': ANY(_lowercase )},
],
[
{'''score''': 0.333, '''label''': ANY(_lowercase )},
{'''score''': 0.333, '''label''': ANY(_lowercase )},
{'''score''': 0.333, '''label''': ANY(_lowercase )},
],
[
{'''score''': 0.333, '''label''': ANY(_lowercase )},
{'''score''': 0.333, '''label''': ANY(_lowercase )},
{'''score''': 0.333, '''label''': ANY(_lowercase )},
],
[
{'''score''': 0.333, '''label''': ANY(_lowercase )},
{'''score''': 0.333, '''label''': ANY(_lowercase )},
{'''score''': 0.333, '''label''': ANY(_lowercase )},
],
[
{'''score''': 0.333, '''label''': ANY(_lowercase )},
{'''score''': 0.333, '''label''': ANY(_lowercase )},
{'''score''': 0.333, '''label''': ANY(_lowercase )},
],
] , )
@require_tf
def a ( self : Optional[Any] ):
__UpperCAmelCase = pipeline(
model='''hf-internal-testing/tiny-random-clip-zero-shot-image-classification''' , framework='''tf''' )
__UpperCAmelCase = Image.open('''./tests/fixtures/tests_samples/COCO/000000039769.png''' )
__UpperCAmelCase = image_classifier(_lowercase , candidate_labels=['''a''', '''b''', '''c'''] )
self.assertEqual(
nested_simplify(_lowercase ) , [{'''score''': 0.333, '''label''': '''a'''}, {'''score''': 0.333, '''label''': '''b'''}, {'''score''': 0.333, '''label''': '''c'''}] , )
__UpperCAmelCase = image_classifier([image] * 5 , candidate_labels=['''A''', '''B''', '''C'''] , batch_size=2 )
self.assertEqual(
nested_simplify(_lowercase ) , [
[
{'''score''': 0.333, '''label''': ANY(_lowercase )},
{'''score''': 0.333, '''label''': ANY(_lowercase )},
{'''score''': 0.333, '''label''': ANY(_lowercase )},
],
[
{'''score''': 0.333, '''label''': ANY(_lowercase )},
{'''score''': 0.333, '''label''': ANY(_lowercase )},
{'''score''': 0.333, '''label''': ANY(_lowercase )},
],
[
{'''score''': 0.333, '''label''': ANY(_lowercase )},
{'''score''': 0.333, '''label''': ANY(_lowercase )},
{'''score''': 0.333, '''label''': ANY(_lowercase )},
],
[
{'''score''': 0.333, '''label''': ANY(_lowercase )},
{'''score''': 0.333, '''label''': ANY(_lowercase )},
{'''score''': 0.333, '''label''': ANY(_lowercase )},
],
[
{'''score''': 0.333, '''label''': ANY(_lowercase )},
{'''score''': 0.333, '''label''': ANY(_lowercase )},
{'''score''': 0.333, '''label''': ANY(_lowercase )},
],
] , )
@slow
@require_torch
def a ( self : List[str] ):
__UpperCAmelCase = pipeline(
task='''zero-shot-image-classification''' , model='''openai/clip-vit-base-patch32''' , )
# This is an image of 2 cats with remotes and no planes
__UpperCAmelCase = Image.open('''./tests/fixtures/tests_samples/COCO/000000039769.png''' )
__UpperCAmelCase = image_classifier(_lowercase , candidate_labels=['''cat''', '''plane''', '''remote'''] )
self.assertEqual(
nested_simplify(_lowercase ) , [
{'''score''': 0.511, '''label''': '''remote'''},
{'''score''': 0.485, '''label''': '''cat'''},
{'''score''': 0.004, '''label''': '''plane'''},
] , )
__UpperCAmelCase = image_classifier([image] * 5 , candidate_labels=['''cat''', '''plane''', '''remote'''] , batch_size=2 )
self.assertEqual(
nested_simplify(_lowercase ) , [
[
{'''score''': 0.511, '''label''': '''remote'''},
{'''score''': 0.485, '''label''': '''cat'''},
{'''score''': 0.004, '''label''': '''plane'''},
],
]
* 5 , )
@slow
@require_tf
def a ( self : Dict ):
__UpperCAmelCase = pipeline(
task='''zero-shot-image-classification''' , model='''openai/clip-vit-base-patch32''' , framework='''tf''' )
# This is an image of 2 cats with remotes and no planes
__UpperCAmelCase = Image.open('''./tests/fixtures/tests_samples/COCO/000000039769.png''' )
__UpperCAmelCase = image_classifier(_lowercase , candidate_labels=['''cat''', '''plane''', '''remote'''] )
self.assertEqual(
nested_simplify(_lowercase ) , [
{'''score''': 0.511, '''label''': '''remote'''},
{'''score''': 0.485, '''label''': '''cat'''},
{'''score''': 0.004, '''label''': '''plane'''},
] , )
__UpperCAmelCase = image_classifier([image] * 5 , candidate_labels=['''cat''', '''plane''', '''remote'''] , batch_size=2 )
self.assertEqual(
nested_simplify(_lowercase ) , [
[
{'''score''': 0.511, '''label''': '''remote'''},
{'''score''': 0.485, '''label''': '''cat'''},
{'''score''': 0.004, '''label''': '''plane'''},
],
]
* 5 , )
| 49 |
from __future__ import annotations
import sys
from collections import deque
from typing import Generic, TypeVar
snake_case__ : Dict = TypeVar("""T""")
class _a ( Generic[T] ):
"""simple docstring"""
A_ = 42 # Cache store of keys
A_ = 42 # References of the keys in cache
A_ = 10 # Maximum capacity of cache
def __init__( self , _UpperCAmelCase ) -> None:
UpperCamelCase_ = deque()
UpperCamelCase_ = set()
if not n:
UpperCamelCase_ = sys.maxsize
elif n < 0:
raise ValueError('n should be an integer greater than 0.' )
else:
UpperCamelCase_ = n
def _UpperCAmelCase ( self , _UpperCAmelCase ) -> None:
if x not in self.key_reference:
if len(self.dq_store ) == LRUCache._MAX_CAPACITY:
UpperCamelCase_ = self.dq_store.pop()
self.key_reference.remove(_UpperCAmelCase )
else:
self.dq_store.remove(_UpperCAmelCase )
self.dq_store.appendleft(_UpperCAmelCase )
self.key_reference.add(_UpperCAmelCase )
def _UpperCAmelCase ( self ) -> None:
for k in self.dq_store:
print(_UpperCAmelCase )
def __repr__( self ) -> str:
return f"""LRUCache({self._MAX_CAPACITY}) => {list(self.dq_store )}"""
if __name__ == "__main__":
import doctest
doctest.testmod()
snake_case__ : LRUCache[str | int] = LRUCache(4)
lru_cache.refer("""A""")
lru_cache.refer(2)
lru_cache.refer(3)
lru_cache.refer("""A""")
lru_cache.refer(4)
lru_cache.refer(5)
lru_cache.display()
print(lru_cache)
assert str(lru_cache) == "LRUCache(4) => [5, 4, 'A', 3]"
| 23 | 0 |
'''simple docstring'''
import gc
import math
import unittest
import torch
from diffusers import UNetaDModel
from diffusers.utils import floats_tensor, logging, slow, torch_all_close, torch_device
from diffusers.utils.testing_utils import enable_full_determinism
from .test_modeling_common import ModelTesterMixin, UNetTesterMixin
UpperCamelCase : Dict = logging.get_logger(__name__)
enable_full_determinism()
class UpperCamelCase__ (a ,a ,unittest.TestCase ):
'''simple docstring'''
_UpperCamelCase = UNetaDModel
_UpperCamelCase = 'sample'
@property
def UpperCamelCase_ ( self ):
lowerCamelCase__ = 4
lowerCamelCase__ = 3
lowerCamelCase__ = (32, 32)
lowerCamelCase__ = floats_tensor((batch_size, num_channels) + sizes ).to(_lowerCAmelCase )
lowerCamelCase__ = torch.tensor([10] ).to(_lowerCAmelCase )
return {"sample": noise, "timestep": time_step}
@property
def UpperCamelCase_ ( self ):
return (3, 32, 32)
@property
def UpperCamelCase_ ( self ):
return (3, 32, 32)
def UpperCamelCase_ ( self ):
lowerCamelCase__ = {
"""block_out_channels""": (32, 64),
"""down_block_types""": ("""DownBlock2D""", """AttnDownBlock2D"""),
"""up_block_types""": ("""AttnUpBlock2D""", """UpBlock2D"""),
"""attention_head_dim""": 3,
"""out_channels""": 3,
"""in_channels""": 3,
"""layers_per_block""": 2,
"""sample_size""": 32,
}
lowerCamelCase__ = self.dummy_input
return init_dict, inputs_dict
class UpperCamelCase__ (a ,a ,unittest.TestCase ):
'''simple docstring'''
_UpperCamelCase = UNetaDModel
_UpperCamelCase = 'sample'
@property
def UpperCamelCase_ ( self ):
lowerCamelCase__ = 4
lowerCamelCase__ = 4
lowerCamelCase__ = (32, 32)
lowerCamelCase__ = floats_tensor((batch_size, num_channels) + sizes ).to(_lowerCAmelCase )
lowerCamelCase__ = torch.tensor([10] ).to(_lowerCAmelCase )
return {"sample": noise, "timestep": time_step}
@property
def UpperCamelCase_ ( self ):
return (4, 32, 32)
@property
def UpperCamelCase_ ( self ):
return (4, 32, 32)
def UpperCamelCase_ ( self ):
lowerCamelCase__ = {
"""sample_size""": 32,
"""in_channels""": 4,
"""out_channels""": 4,
"""layers_per_block""": 2,
"""block_out_channels""": (32, 64),
"""attention_head_dim""": 32,
"""down_block_types""": ("""DownBlock2D""", """DownBlock2D"""),
"""up_block_types""": ("""UpBlock2D""", """UpBlock2D"""),
}
lowerCamelCase__ = self.dummy_input
return init_dict, inputs_dict
def UpperCamelCase_ ( self ):
lowerCamelCase__ , lowerCamelCase__ = UNetaDModel.from_pretrained("""fusing/unet-ldm-dummy-update""" ,output_loading_info=_lowerCAmelCase )
self.assertIsNotNone(_lowerCAmelCase )
self.assertEqual(len(loading_info["""missing_keys"""] ) ,0 )
model.to(_lowerCAmelCase )
lowerCamelCase__ = model(**self.dummy_input ).sample
assert image is not None, "Make sure output is not None"
@unittest.skipIf(torch_device != """cuda""" ,"""This test is supposed to run on GPU""" )
def UpperCamelCase_ ( self ):
lowerCamelCase__ , lowerCamelCase__ = UNetaDModel.from_pretrained("""fusing/unet-ldm-dummy-update""" ,output_loading_info=_lowerCAmelCase )
model.to(_lowerCAmelCase )
lowerCamelCase__ = model(**self.dummy_input ).sample
assert image is not None, "Make sure output is not None"
@unittest.skipIf(torch_device != """cuda""" ,"""This test is supposed to run on GPU""" )
def UpperCamelCase_ ( self ):
# by defautl model loading will use accelerate as `low_cpu_mem_usage=True`
lowerCamelCase__ , lowerCamelCase__ = UNetaDModel.from_pretrained("""fusing/unet-ldm-dummy-update""" ,output_loading_info=_lowerCAmelCase )
model_accelerate.to(_lowerCAmelCase )
model_accelerate.eval()
lowerCamelCase__ = torch.randn(
1 ,model_accelerate.config.in_channels ,model_accelerate.config.sample_size ,model_accelerate.config.sample_size ,generator=torch.manual_seed(0 ) ,)
lowerCamelCase__ = noise.to(_lowerCAmelCase )
lowerCamelCase__ = torch.tensor([10] * noise.shape[0] ).to(_lowerCAmelCase )
lowerCamelCase__ = model_accelerate(_lowerCAmelCase ,_lowerCAmelCase )["""sample"""]
# two models don't need to stay in the device at the same time
del model_accelerate
torch.cuda.empty_cache()
gc.collect()
lowerCamelCase__ , lowerCamelCase__ = UNetaDModel.from_pretrained(
"""fusing/unet-ldm-dummy-update""" ,output_loading_info=_lowerCAmelCase ,low_cpu_mem_usage=_lowerCAmelCase )
model_normal_load.to(_lowerCAmelCase )
model_normal_load.eval()
lowerCamelCase__ = model_normal_load(_lowerCAmelCase ,_lowerCAmelCase )["""sample"""]
assert torch_all_close(_lowerCAmelCase ,_lowerCAmelCase ,rtol=1E-3 )
def UpperCamelCase_ ( self ):
lowerCamelCase__ = UNetaDModel.from_pretrained("""fusing/unet-ldm-dummy-update""" )
model.eval()
model.to(_lowerCAmelCase )
lowerCamelCase__ = torch.randn(
1 ,model.config.in_channels ,model.config.sample_size ,model.config.sample_size ,generator=torch.manual_seed(0 ) ,)
lowerCamelCase__ = noise.to(_lowerCAmelCase )
lowerCamelCase__ = torch.tensor([10] * noise.shape[0] ).to(_lowerCAmelCase )
with torch.no_grad():
lowerCamelCase__ = model(_lowerCAmelCase ,_lowerCAmelCase ).sample
lowerCamelCase__ = output[0, -1, -3:, -3:].flatten().cpu()
# fmt: off
lowerCamelCase__ = torch.tensor([-13.3258, -20.1100, -15.9873, -17.6617, -23.0596, -17.9419, -13.3675, -16.1889, -12.3800] )
# fmt: on
self.assertTrue(torch_all_close(_lowerCAmelCase ,_lowerCAmelCase ,rtol=1E-3 ) )
class UpperCamelCase__ (a ,a ,unittest.TestCase ):
'''simple docstring'''
_UpperCamelCase = UNetaDModel
_UpperCamelCase = 'sample'
@property
def UpperCamelCase_ ( self ,_lowerCAmelCase=(32, 32) ):
lowerCamelCase__ = 4
lowerCamelCase__ = 3
lowerCamelCase__ = floats_tensor((batch_size, num_channels) + sizes ).to(_lowerCAmelCase )
lowerCamelCase__ = torch.tensor(batch_size * [10] ).to(dtype=torch.intaa ,device=_lowerCAmelCase )
return {"sample": noise, "timestep": time_step}
@property
def UpperCamelCase_ ( self ):
return (3, 32, 32)
@property
def UpperCamelCase_ ( self ):
return (3, 32, 32)
def UpperCamelCase_ ( self ):
lowerCamelCase__ = {
"""block_out_channels""": [32, 64, 64, 64],
"""in_channels""": 3,
"""layers_per_block""": 1,
"""out_channels""": 3,
"""time_embedding_type""": """fourier""",
"""norm_eps""": 1E-6,
"""mid_block_scale_factor""": math.sqrt(2.0 ),
"""norm_num_groups""": None,
"""down_block_types""": [
"""SkipDownBlock2D""",
"""AttnSkipDownBlock2D""",
"""SkipDownBlock2D""",
"""SkipDownBlock2D""",
],
"""up_block_types""": [
"""SkipUpBlock2D""",
"""SkipUpBlock2D""",
"""AttnSkipUpBlock2D""",
"""SkipUpBlock2D""",
],
}
lowerCamelCase__ = self.dummy_input
return init_dict, inputs_dict
@slow
def UpperCamelCase_ ( self ):
lowerCamelCase__ , lowerCamelCase__ = UNetaDModel.from_pretrained("""google/ncsnpp-celebahq-256""" ,output_loading_info=_lowerCAmelCase )
self.assertIsNotNone(_lowerCAmelCase )
self.assertEqual(len(loading_info["""missing_keys"""] ) ,0 )
model.to(_lowerCAmelCase )
lowerCamelCase__ = self.dummy_input
lowerCamelCase__ = floats_tensor((4, 3) + (2_56, 2_56) ).to(_lowerCAmelCase )
lowerCamelCase__ = noise
lowerCamelCase__ = model(**_lowerCAmelCase )
assert image is not None, "Make sure output is not None"
@slow
def UpperCamelCase_ ( self ):
lowerCamelCase__ = UNetaDModel.from_pretrained("""google/ncsnpp-celebahq-256""" )
model.to(_lowerCAmelCase )
lowerCamelCase__ = 4
lowerCamelCase__ = 3
lowerCamelCase__ = (2_56, 2_56)
lowerCamelCase__ = torch.ones((batch_size, num_channels) + sizes ).to(_lowerCAmelCase )
lowerCamelCase__ = torch.tensor(batch_size * [1E-4] ).to(_lowerCAmelCase )
with torch.no_grad():
lowerCamelCase__ = model(_lowerCAmelCase ,_lowerCAmelCase ).sample
lowerCamelCase__ = output[0, -3:, -3:, -1].flatten().cpu()
# fmt: off
lowerCamelCase__ = torch.tensor([-4842.8691, -6499.6631, -3800.1953, -7978.2686, -1_0980.7129, -2_0028.8535, 8148.2822, 2342.2905, 567.7608] )
# fmt: on
self.assertTrue(torch_all_close(_lowerCAmelCase ,_lowerCAmelCase ,rtol=1E-2 ) )
def UpperCamelCase_ ( self ):
lowerCamelCase__ = UNetaDModel.from_pretrained("""fusing/ncsnpp-ffhq-ve-dummy-update""" )
model.to(_lowerCAmelCase )
lowerCamelCase__ = 4
lowerCamelCase__ = 3
lowerCamelCase__ = (32, 32)
lowerCamelCase__ = torch.ones((batch_size, num_channels) + sizes ).to(_lowerCAmelCase )
lowerCamelCase__ = torch.tensor(batch_size * [1E-4] ).to(_lowerCAmelCase )
with torch.no_grad():
lowerCamelCase__ = model(_lowerCAmelCase ,_lowerCAmelCase ).sample
lowerCamelCase__ = output[0, -3:, -3:, -1].flatten().cpu()
# fmt: off
lowerCamelCase__ = torch.tensor([-0.0325, -0.0900, -0.0869, -0.0332, -0.0725, -0.0270, -0.0101, 0.0227, 0.0256] )
# fmt: on
self.assertTrue(torch_all_close(_lowerCAmelCase ,_lowerCAmelCase ,rtol=1E-2 ) )
def UpperCamelCase_ ( self ):
# not required for this model
pass
| 50 |
import numpy as np
def _snake_case (__lowercase):
return 1 / (1 + np.exp(-vector))
def _snake_case (__lowercase):
return vector * sigmoid(__lowercase)
if __name__ == "__main__":
import doctest
doctest.testmod()
| 23 | 0 |
'''simple docstring'''
import importlib
import json
import os
from collections import OrderedDict
from typing import Dict, Optional, Union
# Build the list of all image processors
from ...configuration_utils import PretrainedConfig
from ...dynamic_module_utils import get_class_from_dynamic_module, resolve_trust_remote_code
from ...image_processing_utils import ImageProcessingMixin
from ...utils import CONFIG_NAME, IMAGE_PROCESSOR_NAME, get_file_from_repo, logging
from .auto_factory import _LazyAutoMapping
from .configuration_auto import (
CONFIG_MAPPING_NAMES,
AutoConfig,
model_type_to_module_name,
replace_list_option_in_docstrings,
)
a__ : Optional[int] = logging.get_logger(__name__)
a__ : Optional[Any] = OrderedDict(
[
('align', 'EfficientNetImageProcessor'),
('beit', 'BeitImageProcessor'),
('bit', 'BitImageProcessor'),
('blip', 'BlipImageProcessor'),
('blip-2', 'BlipImageProcessor'),
('bridgetower', 'BridgeTowerImageProcessor'),
('chinese_clip', 'ChineseCLIPImageProcessor'),
('clip', 'CLIPImageProcessor'),
('clipseg', 'ViTImageProcessor'),
('conditional_detr', 'ConditionalDetrImageProcessor'),
('convnext', 'ConvNextImageProcessor'),
('convnextv2', 'ConvNextImageProcessor'),
('cvt', 'ConvNextImageProcessor'),
('data2vec-vision', 'BeitImageProcessor'),
('deformable_detr', 'DeformableDetrImageProcessor'),
('deit', 'DeiTImageProcessor'),
('deta', 'DetaImageProcessor'),
('detr', 'DetrImageProcessor'),
('dinat', 'ViTImageProcessor'),
('donut-swin', 'DonutImageProcessor'),
('dpt', 'DPTImageProcessor'),
('efficientformer', 'EfficientFormerImageProcessor'),
('efficientnet', 'EfficientNetImageProcessor'),
('flava', 'FlavaImageProcessor'),
('focalnet', 'BitImageProcessor'),
('git', 'CLIPImageProcessor'),
('glpn', 'GLPNImageProcessor'),
('groupvit', 'CLIPImageProcessor'),
('imagegpt', 'ImageGPTImageProcessor'),
('instructblip', 'BlipImageProcessor'),
('layoutlmv2', 'LayoutLMv2ImageProcessor'),
('layoutlmv3', 'LayoutLMv3ImageProcessor'),
('levit', 'LevitImageProcessor'),
('mask2former', 'Mask2FormerImageProcessor'),
('maskformer', 'MaskFormerImageProcessor'),
('mgp-str', 'ViTImageProcessor'),
('mobilenet_v1', 'MobileNetV1ImageProcessor'),
('mobilenet_v2', 'MobileNetV2ImageProcessor'),
('mobilevit', 'MobileViTImageProcessor'),
('mobilevit', 'MobileViTImageProcessor'),
('mobilevitv2', 'MobileViTImageProcessor'),
('nat', 'ViTImageProcessor'),
('oneformer', 'OneFormerImageProcessor'),
('owlvit', 'OwlViTImageProcessor'),
('perceiver', 'PerceiverImageProcessor'),
('pix2struct', 'Pix2StructImageProcessor'),
('poolformer', 'PoolFormerImageProcessor'),
('regnet', 'ConvNextImageProcessor'),
('resnet', 'ConvNextImageProcessor'),
('sam', 'SamImageProcessor'),
('segformer', 'SegformerImageProcessor'),
('swiftformer', 'ViTImageProcessor'),
('swin', 'ViTImageProcessor'),
('swin2sr', 'Swin2SRImageProcessor'),
('swinv2', 'ViTImageProcessor'),
('table-transformer', 'DetrImageProcessor'),
('timesformer', 'VideoMAEImageProcessor'),
('tvlt', 'TvltImageProcessor'),
('upernet', 'SegformerImageProcessor'),
('van', 'ConvNextImageProcessor'),
('videomae', 'VideoMAEImageProcessor'),
('vilt', 'ViltImageProcessor'),
('vit', 'ViTImageProcessor'),
('vit_hybrid', 'ViTHybridImageProcessor'),
('vit_mae', 'ViTImageProcessor'),
('vit_msn', 'ViTImageProcessor'),
('xclip', 'CLIPImageProcessor'),
('yolos', 'YolosImageProcessor'),
]
)
a__ : Dict = _LazyAutoMapping(CONFIG_MAPPING_NAMES, IMAGE_PROCESSOR_MAPPING_NAMES)
def __snake_case ( SCREAMING_SNAKE_CASE_ : str ) -> Union[str, Any]:
"""simple docstring"""
for module_name, extractors in IMAGE_PROCESSOR_MAPPING_NAMES.items():
if class_name in extractors:
UpperCAmelCase = model_type_to_module_name(SCREAMING_SNAKE_CASE_ )
UpperCAmelCase = importlib.import_module(f".{module_name}" , '''transformers.models''' )
try:
return getattr(SCREAMING_SNAKE_CASE_ , SCREAMING_SNAKE_CASE_ )
except AttributeError:
continue
for _, extractor in IMAGE_PROCESSOR_MAPPING._extra_content.items():
if getattr(SCREAMING_SNAKE_CASE_ , '''__name__''' , SCREAMING_SNAKE_CASE_ ) == class_name:
return extractor
# We did not fine the class, but maybe it's because a dep is missing. In that case, the class will be in the main
# init and we return the proper dummy to get an appropriate error message.
UpperCAmelCase = importlib.import_module('''transformers''' )
if hasattr(SCREAMING_SNAKE_CASE_ , SCREAMING_SNAKE_CASE_ ):
return getattr(SCREAMING_SNAKE_CASE_ , SCREAMING_SNAKE_CASE_ )
return None
def __snake_case ( SCREAMING_SNAKE_CASE_ : Union[str, os.PathLike] , SCREAMING_SNAKE_CASE_ : Optional[Union[str, os.PathLike]] = None , SCREAMING_SNAKE_CASE_ : bool = False , SCREAMING_SNAKE_CASE_ : bool = False , SCREAMING_SNAKE_CASE_ : Optional[Dict[str, str]] = None , SCREAMING_SNAKE_CASE_ : Optional[Union[bool, str]] = None , SCREAMING_SNAKE_CASE_ : Optional[str] = None , SCREAMING_SNAKE_CASE_ : bool = False , **SCREAMING_SNAKE_CASE_ : Union[str, Any] , ) -> List[str]:
"""simple docstring"""
UpperCAmelCase = get_file_from_repo(
SCREAMING_SNAKE_CASE_ , SCREAMING_SNAKE_CASE_ , cache_dir=SCREAMING_SNAKE_CASE_ , force_download=SCREAMING_SNAKE_CASE_ , resume_download=SCREAMING_SNAKE_CASE_ , proxies=SCREAMING_SNAKE_CASE_ , use_auth_token=SCREAMING_SNAKE_CASE_ , revision=SCREAMING_SNAKE_CASE_ , local_files_only=SCREAMING_SNAKE_CASE_ , )
if resolved_config_file is None:
logger.info(
'''Could not locate the image processor configuration file, will try to use the model config instead.''' )
return {}
with open(SCREAMING_SNAKE_CASE_ , encoding='''utf-8''' ) as reader:
return json.load(SCREAMING_SNAKE_CASE_ )
class lowerCAmelCase__ :
'''simple docstring'''
def __init__( self : List[str] ):
raise EnvironmentError(
'''AutoImageProcessor is designed to be instantiated '''
'''using the `AutoImageProcessor.from_pretrained(pretrained_model_name_or_path)` method.''' )
@classmethod
@replace_list_option_in_docstrings(a__ )
def __snake_case ( cls : int , a__ : Union[str, Any] , **a__ : List[str] ):
UpperCAmelCase = kwargs.pop('''config''' , a__ )
UpperCAmelCase = kwargs.pop('''trust_remote_code''' , a__ )
UpperCAmelCase = True
UpperCAmelCase, UpperCAmelCase = ImageProcessingMixin.get_image_processor_dict(a__ , **a__ )
UpperCAmelCase = config_dict.get('''image_processor_type''' , a__ )
UpperCAmelCase = None
if "AutoImageProcessor" in config_dict.get('''auto_map''' , {} ):
UpperCAmelCase = config_dict['''auto_map''']['''AutoImageProcessor''']
# If we still don't have the image processor class, check if we're loading from a previous feature extractor config
# and if so, infer the image processor class from there.
if image_processor_class is None and image_processor_auto_map is None:
UpperCAmelCase = config_dict.pop('''feature_extractor_type''' , a__ )
if feature_extractor_class is not None:
logger.warning(
'''Could not find image processor class in the image processor config or the model config. Loading'''
''' based on pattern matching with the model\'s feature extractor configuration.''' )
UpperCAmelCase = feature_extractor_class.replace('''FeatureExtractor''' , '''ImageProcessor''' )
if "AutoFeatureExtractor" in config_dict.get('''auto_map''' , {} ):
UpperCAmelCase = config_dict['''auto_map''']['''AutoFeatureExtractor''']
UpperCAmelCase = feature_extractor_auto_map.replace('''FeatureExtractor''' , '''ImageProcessor''' )
logger.warning(
'''Could not find image processor auto map in the image processor config or the model config.'''
''' Loading based on pattern matching with the model\'s feature extractor configuration.''' )
# If we don't find the image processor class in the image processor config, let's try the model config.
if image_processor_class is None and image_processor_auto_map is None:
if not isinstance(a__ , a__ ):
UpperCAmelCase = AutoConfig.from_pretrained(a__ , **a__ )
# It could be in `config.image_processor_type``
UpperCAmelCase = getattr(a__ , '''image_processor_type''' , a__ )
if hasattr(a__ , '''auto_map''' ) and "AutoImageProcessor" in config.auto_map:
UpperCAmelCase = config.auto_map['''AutoImageProcessor''']
if image_processor_class is not None:
UpperCAmelCase = image_processor_class_from_name(a__ )
UpperCAmelCase = image_processor_auto_map is not None
UpperCAmelCase = image_processor_class is not None or type(a__ ) in IMAGE_PROCESSOR_MAPPING
UpperCAmelCase = resolve_trust_remote_code(
a__ , a__ , a__ , a__ )
if has_remote_code and trust_remote_code:
UpperCAmelCase = get_class_from_dynamic_module(
a__ , a__ , **a__ )
UpperCAmelCase = kwargs.pop('''code_revision''' , a__ )
if os.path.isdir(a__ ):
image_processor_class.register_for_auto_class()
return image_processor_class.from_dict(a__ , **a__ )
elif image_processor_class is not None:
return image_processor_class.from_dict(a__ , **a__ )
# Last try: we use the IMAGE_PROCESSOR_MAPPING.
elif type(a__ ) in IMAGE_PROCESSOR_MAPPING:
UpperCAmelCase = IMAGE_PROCESSOR_MAPPING[type(a__ )]
return image_processor_class.from_dict(a__ , **a__ )
raise ValueError(
f"Unrecognized image processor in {pretrained_model_name_or_path}. Should have a "
f"`image_processor_type` key in its {IMAGE_PROCESSOR_NAME} of {CONFIG_NAME}, or one of the following "
f"`model_type` keys in its {CONFIG_NAME}: {', '.join(c for c in IMAGE_PROCESSOR_MAPPING_NAMES.keys() )}" )
@staticmethod
def __snake_case ( a__ : Union[str, Any] , a__ : Tuple ):
IMAGE_PROCESSOR_MAPPING.register(a__ , a__ )
| 51 |
import math
from datetime import datetime, timedelta
def _snake_case (__lowercase):
UpperCamelCase_ = year % 19
UpperCamelCase_ = year % 4
UpperCamelCase_ = year % 7
UpperCamelCase_ = math.floor(year / 100)
UpperCamelCase_ = math.floor((13 + 8 * leap_day_inhibits) / 25)
UpperCamelCase_ = leap_day_inhibits / 4
UpperCamelCase_ = (
15 - lunar_orbit_correction + leap_day_inhibits - leap_day_reinstall_number
) % 30
UpperCamelCase_ = (4 + leap_day_inhibits - leap_day_reinstall_number) % 7
# days to be added to March 21
UpperCamelCase_ = (19 * metonic_cycle + secular_moon_shift) % 30
# PHM -> Paschal Full Moon
UpperCamelCase_ = (
2 * julian_leap_year
+ 4 * non_leap_year
+ 6 * days_to_add
+ century_starting_point
) % 7
if days_to_add == 29 and days_from_phm_to_sunday == 6:
return datetime(__lowercase , 4 , 19)
elif days_to_add == 28 and days_from_phm_to_sunday == 6:
return datetime(__lowercase , 4 , 18)
else:
return datetime(__lowercase , 3 , 22) + timedelta(
days=int(days_to_add + days_from_phm_to_sunday))
if __name__ == "__main__":
for year in (1_9_9_4, 2_0_0_0, 2_0_1_0, 2_0_2_1, 2_0_2_3):
snake_case__ : Dict = """will be""" if year > datetime.now().year else """was"""
print(f'Easter in {year} {tense} {gauss_easter(year)}')
| 23 | 0 |
"""simple docstring"""
from math import factorial
A = {str(digit): factorial(digit) for digit in range(10)}
def __A ( a_ :int) -> int:
if not isinstance(a_ , a_):
raise TypeError('''Parameter number must be int''')
if number < 0:
raise ValueError('''Parameter number must be greater than or equal to 0''')
# Converts number in string to iterate on its digits and adds its factorial.
return sum(DIGIT_FACTORIAL[digit] for digit in str(a_))
def __A ( a_ :int = 60 , a_ :int = 1_00_00_00) -> int:
if not isinstance(a_ , a_) or not isinstance(a_ , a_):
raise TypeError('''Parameters chain_length and number_limit must be int''')
if chain_length <= 0 or number_limit <= 0:
raise ValueError(
'''Parameters chain_length and number_limit must be greater than 0''')
# the counter for the chains with the exact desired length
__a : int = 0
# the cached sizes of the previous chains
__a : dict[int, int] = {}
for start_chain_element in range(1 , a_):
# The temporary set will contain the elements of the chain
__a : Tuple = set()
__a : Union[str, Any] = 0
# Stop computing the chain when you find a cached size, a repeating item or the
# length is greater then the desired one.
__a : str = start_chain_element
while (
chain_element not in chain_sets_lengths
and chain_element not in chain_set
and chain_set_length <= chain_length
):
chain_set.add(a_)
chain_set_length += 1
__a : Optional[int] = digit_factorial_sum(a_)
if chain_element in chain_sets_lengths:
chain_set_length += chain_sets_lengths[chain_element]
__a : Optional[int] = chain_set_length
# If chain contains the exact amount of elements increase the counter
if chain_set_length == chain_length:
chains_counter += 1
return chains_counter
if __name__ == "__main__":
import doctest
doctest.testmod()
print(F'{solution()}') | 52 |
import requests
def _snake_case (__lowercase , __lowercase):
UpperCamelCase_ = {'Content-Type': 'application/json'}
UpperCamelCase_ = requests.post(__lowercase , json={'text': message_body} , headers=__lowercase)
if response.status_code != 200:
UpperCamelCase_ = (
'Request to slack returned an error '
f"""{response.status_code}, the response is:\n{response.text}"""
)
raise ValueError(__lowercase)
if __name__ == "__main__":
# Set the slack url to the one provided by Slack when you create the webhook at
# https://my.slack.com/services/new/incoming-webhook/
send_slack_message("""<YOUR MESSAGE BODY>""", """<SLACK CHANNEL URL>""")
| 23 | 0 |
import unittest
import numpy as np
import requests
from transformers.testing_utils import require_torch, require_vision
from transformers.utils import is_torch_available, is_vision_available
from ...test_image_processing_common import ImageProcessingSavingTestMixin, prepare_image_inputs
if is_torch_available():
import torch
from transformers.pytorch_utils import is_torch_greater_or_equal_than_1_11
else:
_snake_case : Dict = False
if is_vision_available():
from PIL import Image
from transformers import PixaStructImageProcessor
class _UpperCAmelCase ( unittest.TestCase ):
"""simple docstring"""
def __init__( self : Optional[int] , lowerCAmelCase_ : List[Any] , lowerCAmelCase_ : List[str]=7 , lowerCAmelCase_ : Dict=3 , lowerCAmelCase_ : Optional[Any]=1_8 , lowerCAmelCase_ : Optional[Any]=3_0 , lowerCAmelCase_ : str=4_0_0 , lowerCAmelCase_ : Union[str, Any]=None , lowerCAmelCase_ : Optional[int]=True , lowerCAmelCase_ : str=True , lowerCAmelCase_ : str=None , ) -> Tuple:
__lowerCAmelCase = size if size is not None else {'height': 2_0, 'width': 2_0}
__lowerCAmelCase = parent
__lowerCAmelCase = batch_size
__lowerCAmelCase = num_channels
__lowerCAmelCase = image_size
__lowerCAmelCase = min_resolution
__lowerCAmelCase = max_resolution
__lowerCAmelCase = size
__lowerCAmelCase = do_normalize
__lowerCAmelCase = do_convert_rgb
__lowerCAmelCase = [5_1_2, 1_0_2_4, 2_0_4_8, 4_0_9_6]
__lowerCAmelCase = patch_size if patch_size is not None else {'height': 1_6, 'width': 1_6}
def lowercase ( self : List[Any] ) -> Tuple:
return {"do_normalize": self.do_normalize, "do_convert_rgb": self.do_convert_rgb}
def lowercase ( self : Union[str, Any] ) -> Union[str, Any]:
__lowerCAmelCase = 'https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/transformers/tasks/australia.jpg'
__lowerCAmelCase = Image.open(requests.get(lowerCAmelCase_ , stream=lowerCAmelCase_ ).raw ).convert('RGB' )
return raw_image
@unittest.skipIf(
not is_torch_greater_or_equal_than_1_11 , reason="""`Pix2StructImageProcessor` requires `torch>=1.11.0`.""" , )
@require_torch
@require_vision
class _UpperCAmelCase ( _UpperCamelCase , unittest.TestCase ):
"""simple docstring"""
a_ = PixaStructImageProcessor if is_vision_available() else None
def lowercase ( self : List[Any] ) -> Dict:
__lowerCAmelCase = PixaStructImageProcessingTester(self )
@property
def lowercase ( self : List[str] ) -> Any:
return self.image_processor_tester.prepare_image_processor_dict()
def lowercase ( self : Union[str, Any] ) -> Union[str, Any]:
__lowerCAmelCase = self.image_processing_class(**self.image_processor_dict )
self.assertTrue(hasattr(lowerCAmelCase_ , 'do_normalize' ) )
self.assertTrue(hasattr(lowerCAmelCase_ , 'do_convert_rgb' ) )
def lowercase ( self : str ) -> Dict:
__lowerCAmelCase = self.image_processor_tester.prepare_dummy_image()
__lowerCAmelCase = self.image_processing_class(**self.image_processor_dict )
__lowerCAmelCase = 2_0_4_8
__lowerCAmelCase = image_processor(lowerCAmelCase_ , return_tensors='pt' , max_patches=lowerCAmelCase_ )
self.assertTrue(torch.allclose(inputs.flattened_patches.mean() , torch.tensor(0.06_06 ) , atol=1e-3 , rtol=1e-3 ) )
def lowercase ( self : Any ) -> Dict:
# Initialize image_processor
__lowerCAmelCase = self.image_processing_class(**self.image_processor_dict )
# create random PIL images
__lowerCAmelCase = prepare_image_inputs(self.image_processor_tester , equal_resolution=lowerCAmelCase_ )
for image in image_inputs:
self.assertIsInstance(lowerCAmelCase_ , Image.Image )
# Test not batched input
__lowerCAmelCase = (
(self.image_processor_tester.patch_size['height'] * self.image_processor_tester.patch_size['width'])
* self.image_processor_tester.num_channels
) + 2
for max_patch in self.image_processor_tester.max_patches:
# Test not batched input
__lowerCAmelCase = image_processor(
image_inputs[0] , return_tensors='pt' , max_patches=lowerCAmelCase_ ).flattened_patches
self.assertEqual(
encoded_images.shape , (1, max_patch, expected_hidden_dim) , )
# Test batched
__lowerCAmelCase = image_processor(
lowerCAmelCase_ , return_tensors='pt' , max_patches=lowerCAmelCase_ ).flattened_patches
self.assertEqual(
encoded_images.shape , (self.image_processor_tester.batch_size, max_patch, expected_hidden_dim) , )
def lowercase ( self : List[str] ) -> Union[str, Any]:
# Initialize image_processor
__lowerCAmelCase = self.image_processing_class(**self.image_processor_dict )
# create random PIL images
__lowerCAmelCase = prepare_image_inputs(self.image_processor_tester , equal_resolution=lowerCAmelCase_ )
for image in image_inputs:
self.assertIsInstance(lowerCAmelCase_ , Image.Image )
# Test not batched input
__lowerCAmelCase = (
(self.image_processor_tester.patch_size['height'] * self.image_processor_tester.patch_size['width'])
* self.image_processor_tester.num_channels
) + 2
__lowerCAmelCase = True
for max_patch in self.image_processor_tester.max_patches:
# Test not batched input
with self.assertRaises(lowerCAmelCase_ ):
__lowerCAmelCase = image_processor(
image_inputs[0] , return_tensors='pt' , max_patches=lowerCAmelCase_ ).flattened_patches
__lowerCAmelCase = 'Hello'
__lowerCAmelCase = image_processor(
image_inputs[0] , return_tensors='pt' , max_patches=lowerCAmelCase_ , header_text=lowerCAmelCase_ ).flattened_patches
self.assertEqual(
encoded_images.shape , (1, max_patch, expected_hidden_dim) , )
# Test batched
__lowerCAmelCase = image_processor(
lowerCAmelCase_ , return_tensors='pt' , max_patches=lowerCAmelCase_ , header_text=lowerCAmelCase_ ).flattened_patches
self.assertEqual(
encoded_images.shape , (self.image_processor_tester.batch_size, max_patch, expected_hidden_dim) , )
def lowercase ( self : Any ) -> Tuple:
# Initialize image_processor
__lowerCAmelCase = self.image_processing_class(**self.image_processor_dict )
# create random numpy tensors
__lowerCAmelCase = prepare_image_inputs(self.image_processor_tester , equal_resolution=lowerCAmelCase_ , numpify=lowerCAmelCase_ )
for image in image_inputs:
self.assertIsInstance(lowerCAmelCase_ , np.ndarray )
__lowerCAmelCase = (
(self.image_processor_tester.patch_size['height'] * self.image_processor_tester.patch_size['width'])
* self.image_processor_tester.num_channels
) + 2
for max_patch in self.image_processor_tester.max_patches:
# Test not batched input
__lowerCAmelCase = image_processor(
image_inputs[0] , return_tensors='pt' , max_patches=lowerCAmelCase_ ).flattened_patches
self.assertEqual(
encoded_images.shape , (1, max_patch, expected_hidden_dim) , )
# Test batched
__lowerCAmelCase = image_processor(
lowerCAmelCase_ , return_tensors='pt' , max_patches=lowerCAmelCase_ ).flattened_patches
self.assertEqual(
encoded_images.shape , (self.image_processor_tester.batch_size, max_patch, expected_hidden_dim) , )
def lowercase ( self : Optional[Any] ) -> List[Any]:
# Initialize image_processor
__lowerCAmelCase = self.image_processing_class(**self.image_processor_dict )
# create random PyTorch tensors
__lowerCAmelCase = prepare_image_inputs(self.image_processor_tester , equal_resolution=lowerCAmelCase_ , torchify=lowerCAmelCase_ )
for image in image_inputs:
self.assertIsInstance(lowerCAmelCase_ , torch.Tensor )
# Test not batched input
__lowerCAmelCase = (
(self.image_processor_tester.patch_size['height'] * self.image_processor_tester.patch_size['width'])
* self.image_processor_tester.num_channels
) + 2
for max_patch in self.image_processor_tester.max_patches:
# Test not batched input
__lowerCAmelCase = image_processor(
image_inputs[0] , return_tensors='pt' , max_patches=lowerCAmelCase_ ).flattened_patches
self.assertEqual(
encoded_images.shape , (1, max_patch, expected_hidden_dim) , )
# Test batched
__lowerCAmelCase = image_processor(
lowerCAmelCase_ , return_tensors='pt' , max_patches=lowerCAmelCase_ ).flattened_patches
self.assertEqual(
encoded_images.shape , (self.image_processor_tester.batch_size, max_patch, expected_hidden_dim) , )
@unittest.skipIf(
not is_torch_greater_or_equal_than_1_11 , reason="""`Pix2StructImageProcessor` requires `torch>=1.11.0`.""" , )
@require_torch
@require_vision
class _UpperCAmelCase ( _UpperCamelCase , unittest.TestCase ):
"""simple docstring"""
a_ = PixaStructImageProcessor if is_vision_available() else None
def lowercase ( self : Any ) -> int:
__lowerCAmelCase = PixaStructImageProcessingTester(self , num_channels=4 )
__lowerCAmelCase = 3
@property
def lowercase ( self : Any ) -> Optional[int]:
return self.image_processor_tester.prepare_image_processor_dict()
def lowercase ( self : Dict ) -> Tuple:
__lowerCAmelCase = self.image_processing_class(**self.image_processor_dict )
self.assertTrue(hasattr(lowerCAmelCase_ , 'do_normalize' ) )
self.assertTrue(hasattr(lowerCAmelCase_ , 'do_convert_rgb' ) )
def lowercase ( self : List[Any] ) -> Any:
# Initialize image_processor
__lowerCAmelCase = self.image_processing_class(**self.image_processor_dict )
# create random PIL images
__lowerCAmelCase = prepare_image_inputs(self.image_processor_tester , equal_resolution=lowerCAmelCase_ )
for image in image_inputs:
self.assertIsInstance(lowerCAmelCase_ , Image.Image )
# Test not batched input
__lowerCAmelCase = (
(self.image_processor_tester.patch_size['height'] * self.image_processor_tester.patch_size['width'])
* (self.image_processor_tester.num_channels - 1)
) + 2
for max_patch in self.image_processor_tester.max_patches:
# Test not batched input
__lowerCAmelCase = image_processor(
image_inputs[0] , return_tensors='pt' , max_patches=lowerCAmelCase_ ).flattened_patches
self.assertEqual(
encoded_images.shape , (1, max_patch, expected_hidden_dim) , )
# Test batched
__lowerCAmelCase = image_processor(
lowerCAmelCase_ , return_tensors='pt' , max_patches=lowerCAmelCase_ ).flattened_patches
self.assertEqual(
encoded_images.shape , (self.image_processor_tester.batch_size, max_patch, expected_hidden_dim) , )
| 53 |
import re
from pathlib import Path
from unittest import TestCase
import pytest
@pytest.mark.integration
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
def _UpperCAmelCase ( self , _UpperCAmelCase ) -> Dict:
with open(_UpperCAmelCase , encoding='utf-8' ) as input_file:
UpperCamelCase_ = re.compile(R'(?!.*\b(?:encoding|rb|w|wb|w+|wb+|ab|ab+)\b)(?<=\s)(open)\((.*)\)' )
UpperCamelCase_ = input_file.read()
UpperCamelCase_ = regexp.search(_UpperCAmelCase )
return match
def _UpperCAmelCase ( self , _UpperCAmelCase ) -> Dict:
with open(_UpperCAmelCase , encoding='utf-8' ) as input_file:
UpperCamelCase_ = re.compile(R'#[^\r\n]*print\(|\"[^\r\n]*print\(|\"\"\".*?print\(.*?\"\"\"|(print\()' , re.DOTALL )
UpperCamelCase_ = input_file.read()
# use `re.finditer` to handle the case where the ignored groups would be matched first by `re.search`
UpperCamelCase_ = regexp.finditer(_UpperCAmelCase )
UpperCamelCase_ = [match for match in matches if match is not None and match.group(1 ) is not None]
return matches[0] if matches else None
def _UpperCAmelCase ( self ) -> List[str]:
UpperCamelCase_ = Path('./datasets' )
UpperCamelCase_ = list(dataset_paths.absolute().glob('**/*.py' ) )
for dataset in dataset_files:
if self._no_encoding_on_file_open(str(_UpperCAmelCase ) ):
raise AssertionError(f"""open(...) must use utf-8 encoding in {dataset}""" )
def _UpperCAmelCase ( self ) -> Optional[int]:
UpperCamelCase_ = Path('./datasets' )
UpperCamelCase_ = list(dataset_paths.absolute().glob('**/*.py' ) )
for dataset in dataset_files:
if self._no_print_statements(str(_UpperCAmelCase ) ):
raise AssertionError(f"""print statement found in {dataset}. Use datasets.logger/logging instead.""" )
| 23 | 0 |
import inspect
import unittest
from datasets import load_dataset
from packaging import version
from transformers import BeitConfig
from transformers.models.auto import get_values
from transformers.testing_utils import require_torch, require_torch_multi_gpu, require_vision, slow, torch_device
from transformers.utils import cached_property, is_torch_available, is_vision_available
from ...test_configuration_common import ConfigTester
from ...test_modeling_common import ModelTesterMixin, _config_zero_init, floats_tensor, ids_tensor
from ...test_pipeline_mixin import PipelineTesterMixin
if is_torch_available():
import torch
from torch import nn
from transformers import (
MODEL_MAPPING,
BeitForImageClassification,
BeitForMaskedImageModeling,
BeitForSemanticSegmentation,
BeitModel,
)
from transformers.models.beit.modeling_beit import BEIT_PRETRAINED_MODEL_ARCHIVE_LIST
if is_vision_available():
import PIL
from PIL import Image
from transformers import BeitImageProcessor
class A :
def __init__( self: Any , _lowerCAmelCase: Tuple , _lowerCAmelCase: List[Any]=100 , _lowerCAmelCase: Optional[Any]=13 , _lowerCAmelCase: str=30 , _lowerCAmelCase: str=2 , _lowerCAmelCase: str=3 , _lowerCAmelCase: List[str]=True , _lowerCAmelCase: Union[str, Any]=True , _lowerCAmelCase: List[str]=32 , _lowerCAmelCase: str=4 , _lowerCAmelCase: List[str]=4 , _lowerCAmelCase: str=37 , _lowerCAmelCase: str="gelu" , _lowerCAmelCase: Tuple=0.1 , _lowerCAmelCase: Optional[Any]=0.1 , _lowerCAmelCase: str=10 , _lowerCAmelCase: Optional[int]=0.02 , _lowerCAmelCase: Optional[Any]=3 , _lowerCAmelCase: Dict=None , _lowerCAmelCase: Any=[0, 1, 2, 3] , ) -> Optional[int]:
'''simple docstring'''
UpperCAmelCase_ =parent
UpperCAmelCase_ =100
UpperCAmelCase_ =batch_size
UpperCAmelCase_ =image_size
UpperCAmelCase_ =patch_size
UpperCAmelCase_ =num_channels
UpperCAmelCase_ =is_training
UpperCAmelCase_ =use_labels
UpperCAmelCase_ =hidden_size
UpperCAmelCase_ =num_hidden_layers
UpperCAmelCase_ =num_attention_heads
UpperCAmelCase_ =intermediate_size
UpperCAmelCase_ =hidden_act
UpperCAmelCase_ =hidden_dropout_prob
UpperCAmelCase_ =attention_probs_dropout_prob
UpperCAmelCase_ =type_sequence_label_size
UpperCAmelCase_ =initializer_range
UpperCAmelCase_ =scope
UpperCAmelCase_ =out_indices
UpperCAmelCase_ =num_labels
# in BeiT, the seq length equals the number of patches + 1 (we add 1 for the [CLS] token)
UpperCAmelCase_ =(image_size // patch_size) ** 2
UpperCAmelCase_ =num_patches + 1
def lowerCAmelCase__ ( self: Optional[int] ) -> Any:
'''simple docstring'''
UpperCAmelCase_ =floats_tensor([self.batch_size, self.num_channels, self.image_size, self.image_size] )
UpperCAmelCase_ =None
UpperCAmelCase_ =None
if self.use_labels:
UpperCAmelCase_ =ids_tensor([self.batch_size] , self.type_sequence_label_size )
UpperCAmelCase_ =ids_tensor([self.batch_size, self.image_size, self.image_size] , self.num_labels )
UpperCAmelCase_ =self.get_config()
return config, pixel_values, labels, pixel_labels
def lowerCAmelCase__ ( self: Optional[Any] ) -> List[Any]:
'''simple docstring'''
return BeitConfig(
vocab_size=self.vocab_size , image_size=self.image_size , patch_size=self.patch_size , num_channels=self.num_channels , hidden_size=self.hidden_size , num_hidden_layers=self.num_hidden_layers , num_attention_heads=self.num_attention_heads , intermediate_size=self.intermediate_size , hidden_act=self.hidden_act , hidden_dropout_prob=self.hidden_dropout_prob , attention_probs_dropout_prob=self.attention_probs_dropout_prob , is_decoder=_lowerCAmelCase , initializer_range=self.initializer_range , out_indices=self.out_indices , )
def lowerCAmelCase__ ( self: Union[str, Any] , _lowerCAmelCase: Tuple , _lowerCAmelCase: int , _lowerCAmelCase: Union[str, Any] , _lowerCAmelCase: Union[str, Any] ) -> int:
'''simple docstring'''
UpperCAmelCase_ =BeitModel(config=_lowerCAmelCase )
model.to(_lowerCAmelCase )
model.eval()
UpperCAmelCase_ =model(_lowerCAmelCase )
self.parent.assertEqual(result.last_hidden_state.shape , (self.batch_size, self.seq_length, self.hidden_size) )
def lowerCAmelCase__ ( self: List[str] , _lowerCAmelCase: Union[str, Any] , _lowerCAmelCase: Union[str, Any] , _lowerCAmelCase: Union[str, Any] , _lowerCAmelCase: Dict ) -> Dict:
'''simple docstring'''
UpperCAmelCase_ =BeitForMaskedImageModeling(config=_lowerCAmelCase )
model.to(_lowerCAmelCase )
model.eval()
UpperCAmelCase_ =model(_lowerCAmelCase )
self.parent.assertEqual(result.logits.shape , (self.batch_size, self.seq_length - 1, self.vocab_size) )
def lowerCAmelCase__ ( self: Optional[Any] , _lowerCAmelCase: int , _lowerCAmelCase: Optional[Any] , _lowerCAmelCase: List[str] , _lowerCAmelCase: Optional[Any] ) -> Optional[Any]:
'''simple docstring'''
UpperCAmelCase_ =self.type_sequence_label_size
UpperCAmelCase_ =BeitForImageClassification(_lowerCAmelCase )
model.to(_lowerCAmelCase )
model.eval()
UpperCAmelCase_ =model(_lowerCAmelCase , labels=_lowerCAmelCase )
self.parent.assertEqual(result.logits.shape , (self.batch_size, self.type_sequence_label_size) )
# test greyscale images
UpperCAmelCase_ =1
UpperCAmelCase_ =BeitForImageClassification(_lowerCAmelCase )
model.to(_lowerCAmelCase )
model.eval()
UpperCAmelCase_ =floats_tensor([self.batch_size, 1, self.image_size, self.image_size] )
UpperCAmelCase_ =model(_lowerCAmelCase , labels=_lowerCAmelCase )
self.parent.assertEqual(result.logits.shape , (self.batch_size, self.type_sequence_label_size) )
def lowerCAmelCase__ ( self: List[str] , _lowerCAmelCase: str , _lowerCAmelCase: int , _lowerCAmelCase: Any , _lowerCAmelCase: int ) -> str:
'''simple docstring'''
UpperCAmelCase_ =self.num_labels
UpperCAmelCase_ =BeitForSemanticSegmentation(_lowerCAmelCase )
model.to(_lowerCAmelCase )
model.eval()
UpperCAmelCase_ =model(_lowerCAmelCase )
self.parent.assertEqual(
result.logits.shape , (self.batch_size, self.num_labels, self.image_size * 2, self.image_size * 2) )
UpperCAmelCase_ =model(_lowerCAmelCase , labels=_lowerCAmelCase )
self.parent.assertEqual(
result.logits.shape , (self.batch_size, self.num_labels, self.image_size * 2, self.image_size * 2) )
def lowerCAmelCase__ ( self: Any ) -> Union[str, Any]:
'''simple docstring'''
UpperCAmelCase_ =self.prepare_config_and_inputs()
UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ =config_and_inputs
UpperCAmelCase_ ={"pixel_values": pixel_values}
return config, inputs_dict
@require_torch
class A ( __lowercase , __lowercase , unittest.TestCase ):
_snake_case =(
(BeitModel, BeitForImageClassification, BeitForMaskedImageModeling, BeitForSemanticSegmentation)
if is_torch_available()
else ()
)
_snake_case =(
{
'''feature-extraction''': BeitModel,
'''image-classification''': BeitForImageClassification,
'''image-segmentation''': BeitForSemanticSegmentation,
}
if is_torch_available()
else {}
)
_snake_case =False
_snake_case =False
_snake_case =False
def lowerCAmelCase__ ( self: Union[str, Any] ) -> List[str]:
'''simple docstring'''
UpperCAmelCase_ =BeitModelTester(self )
UpperCAmelCase_ =ConfigTester(self , config_class=_lowerCAmelCase , has_text_modality=_lowerCAmelCase , hidden_size=37 )
def lowerCAmelCase__ ( self: List[Any] ) -> int:
'''simple docstring'''
self.config_tester.run_common_tests()
@unittest.skip(reason="BEiT does not use inputs_embeds" )
def lowerCAmelCase__ ( self: Union[str, Any] ) -> Any:
'''simple docstring'''
pass
@require_torch_multi_gpu
@unittest.skip(reason="BEiT has some layers using `add_module` which doesn't work well with `nn.DataParallel`" )
def lowerCAmelCase__ ( self: List[Any] ) -> Any:
'''simple docstring'''
pass
def lowerCAmelCase__ ( self: str ) -> Optional[Any]:
'''simple docstring'''
UpperCAmelCase_ , UpperCAmelCase_ =self.model_tester.prepare_config_and_inputs_for_common()
for model_class in self.all_model_classes:
UpperCAmelCase_ =model_class(_lowerCAmelCase )
self.assertIsInstance(model.get_input_embeddings() , (nn.Module) )
UpperCAmelCase_ =model.get_output_embeddings()
self.assertTrue(x is None or isinstance(_lowerCAmelCase , nn.Linear ) )
def lowerCAmelCase__ ( self: str ) -> List[str]:
'''simple docstring'''
UpperCAmelCase_ , UpperCAmelCase_ =self.model_tester.prepare_config_and_inputs_for_common()
for model_class in self.all_model_classes:
UpperCAmelCase_ =model_class(_lowerCAmelCase )
UpperCAmelCase_ =inspect.signature(model.forward )
# signature.parameters is an OrderedDict => so arg_names order is deterministic
UpperCAmelCase_ =[*signature.parameters.keys()]
UpperCAmelCase_ =["pixel_values"]
self.assertListEqual(arg_names[:1] , _lowerCAmelCase )
def lowerCAmelCase__ ( self: Any ) -> int:
'''simple docstring'''
UpperCAmelCase_ =self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_model(*_lowerCAmelCase )
def lowerCAmelCase__ ( self: str ) -> Tuple:
'''simple docstring'''
UpperCAmelCase_ =self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_for_masked_lm(*_lowerCAmelCase )
def lowerCAmelCase__ ( self: List[Any] ) -> Any:
'''simple docstring'''
UpperCAmelCase_ =self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_for_image_classification(*_lowerCAmelCase )
def lowerCAmelCase__ ( self: Union[str, Any] ) -> Optional[int]:
'''simple docstring'''
UpperCAmelCase_ =self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_for_semantic_segmentation(*_lowerCAmelCase )
def lowerCAmelCase__ ( self: Optional[int] ) -> Union[str, Any]:
'''simple docstring'''
if not self.model_tester.is_training:
return
UpperCAmelCase_ , UpperCAmelCase_ =self.model_tester.prepare_config_and_inputs_for_common()
UpperCAmelCase_ =True
for model_class in self.all_model_classes:
# we don't test BeitForMaskedImageModeling
if model_class in [*get_values(_lowerCAmelCase ), BeitForMaskedImageModeling]:
continue
UpperCAmelCase_ =model_class(_lowerCAmelCase )
model.to(_lowerCAmelCase )
model.train()
UpperCAmelCase_ =self._prepare_for_class(_lowerCAmelCase , _lowerCAmelCase , return_labels=_lowerCAmelCase )
UpperCAmelCase_ =model(**_lowerCAmelCase ).loss
loss.backward()
def lowerCAmelCase__ ( self: List[Any] ) -> Any:
'''simple docstring'''
UpperCAmelCase_ , UpperCAmelCase_ =self.model_tester.prepare_config_and_inputs_for_common()
if not self.model_tester.is_training:
return
UpperCAmelCase_ =False
UpperCAmelCase_ =True
for model_class in self.all_model_classes:
# we don't test BeitForMaskedImageModeling
if (
model_class in [*get_values(_lowerCAmelCase ), BeitForMaskedImageModeling]
or not model_class.supports_gradient_checkpointing
):
continue
UpperCAmelCase_ =model_class(_lowerCAmelCase )
model.gradient_checkpointing_enable()
model.to(_lowerCAmelCase )
model.train()
UpperCAmelCase_ =self._prepare_for_class(_lowerCAmelCase , _lowerCAmelCase , return_labels=_lowerCAmelCase )
UpperCAmelCase_ =model(**_lowerCAmelCase ).loss
loss.backward()
def lowerCAmelCase__ ( self: Tuple ) -> Dict:
'''simple docstring'''
UpperCAmelCase_ , UpperCAmelCase_ =self.model_tester.prepare_config_and_inputs_for_common()
UpperCAmelCase_ =_config_zero_init(_lowerCAmelCase )
for model_class in self.all_model_classes:
UpperCAmelCase_ =model_class(config=_lowerCAmelCase )
for name, param in model.named_parameters():
# we skip lambda parameters as these require special initial values
# determined by config.layer_scale_init_value
if "lambda" in name:
continue
if param.requires_grad:
self.assertIn(
((param.data.mean() * 1e9).round() / 1e9).item() , [0.0, 1.0] , msg=F'Parameter {name} of model {model_class} seems not properly initialized' , )
@slow
def lowerCAmelCase__ ( self: Optional[Any] ) -> List[Any]:
'''simple docstring'''
for model_name in BEIT_PRETRAINED_MODEL_ARCHIVE_LIST[:1]:
UpperCAmelCase_ =BeitModel.from_pretrained(_lowerCAmelCase )
self.assertIsNotNone(_lowerCAmelCase )
def a__ ( ):
'''simple docstring'''
UpperCAmelCase_ =Image.open("./tests/fixtures/tests_samples/COCO/000000039769.png" )
return image
@require_torch
@require_vision
class A ( unittest.TestCase ):
@cached_property
def lowerCAmelCase__ ( self: Union[str, Any] ) -> List[Any]:
'''simple docstring'''
return BeitImageProcessor.from_pretrained("microsoft/beit-base-patch16-224" ) if is_vision_available() else None
@slow
def lowerCAmelCase__ ( self: Union[str, Any] ) -> int:
'''simple docstring'''
UpperCAmelCase_ =BeitForMaskedImageModeling.from_pretrained("microsoft/beit-base-patch16-224-pt22k" ).to(_lowerCAmelCase )
UpperCAmelCase_ =self.default_image_processor
UpperCAmelCase_ =prepare_img()
UpperCAmelCase_ =image_processor(images=_lowerCAmelCase , return_tensors="pt" ).pixel_values.to(_lowerCAmelCase )
# prepare bool_masked_pos
UpperCAmelCase_ =torch.ones((1, 196) , dtype=torch.bool ).to(_lowerCAmelCase )
# forward pass
with torch.no_grad():
UpperCAmelCase_ =model(pixel_values=_lowerCAmelCase , bool_masked_pos=_lowerCAmelCase )
UpperCAmelCase_ =outputs.logits
# verify the logits
UpperCAmelCase_ =torch.Size((1, 196, 8192) )
self.assertEqual(logits.shape , _lowerCAmelCase )
UpperCAmelCase_ =torch.tensor(
[[-3.24_37, 0.50_72, -13.91_74], [-3.24_56, 0.49_48, -13.94_01], [-3.20_33, 0.51_21, -13.85_50]] ).to(_lowerCAmelCase )
self.assertTrue(torch.allclose(logits[bool_masked_pos][:3, :3] , _lowerCAmelCase , atol=1e-2 ) )
@slow
def lowerCAmelCase__ ( self: Optional[int] ) -> Any:
'''simple docstring'''
UpperCAmelCase_ =BeitForImageClassification.from_pretrained("microsoft/beit-base-patch16-224" ).to(_lowerCAmelCase )
UpperCAmelCase_ =self.default_image_processor
UpperCAmelCase_ =prepare_img()
UpperCAmelCase_ =image_processor(images=_lowerCAmelCase , return_tensors="pt" ).to(_lowerCAmelCase )
# forward pass
with torch.no_grad():
UpperCAmelCase_ =model(**_lowerCAmelCase )
UpperCAmelCase_ =outputs.logits
# verify the logits
UpperCAmelCase_ =torch.Size((1, 1000) )
self.assertEqual(logits.shape , _lowerCAmelCase )
UpperCAmelCase_ =torch.tensor([-1.23_85, -1.09_87, -1.01_08] ).to(_lowerCAmelCase )
self.assertTrue(torch.allclose(logits[0, :3] , _lowerCAmelCase , atol=1e-4 ) )
UpperCAmelCase_ =281
self.assertEqual(logits.argmax(-1 ).item() , _lowerCAmelCase )
@slow
def lowerCAmelCase__ ( self: str ) -> Optional[int]:
'''simple docstring'''
UpperCAmelCase_ =BeitForImageClassification.from_pretrained("microsoft/beit-large-patch16-224-pt22k-ft22k" ).to(
_lowerCAmelCase )
UpperCAmelCase_ =self.default_image_processor
UpperCAmelCase_ =prepare_img()
UpperCAmelCase_ =image_processor(images=_lowerCAmelCase , return_tensors="pt" ).to(_lowerCAmelCase )
# forward pass
with torch.no_grad():
UpperCAmelCase_ =model(**_lowerCAmelCase )
UpperCAmelCase_ =outputs.logits
# verify the logits
UpperCAmelCase_ =torch.Size((1, 2_1841) )
self.assertEqual(logits.shape , _lowerCAmelCase )
UpperCAmelCase_ =torch.tensor([1.68_81, -0.27_87, 0.59_01] ).to(_lowerCAmelCase )
self.assertTrue(torch.allclose(logits[0, :3] , _lowerCAmelCase , atol=1e-4 ) )
UpperCAmelCase_ =2396
self.assertEqual(logits.argmax(-1 ).item() , _lowerCAmelCase )
@slow
def lowerCAmelCase__ ( self: str ) -> str:
'''simple docstring'''
UpperCAmelCase_ =BeitForSemanticSegmentation.from_pretrained("microsoft/beit-base-finetuned-ade-640-640" )
UpperCAmelCase_ =model.to(_lowerCAmelCase )
UpperCAmelCase_ =BeitImageProcessor(do_resize=_lowerCAmelCase , size=640 , do_center_crop=_lowerCAmelCase )
UpperCAmelCase_ =load_dataset("hf-internal-testing/fixtures_ade20k" , split="test" )
UpperCAmelCase_ =Image.open(ds[0]["file"] )
UpperCAmelCase_ =image_processor(images=_lowerCAmelCase , return_tensors="pt" ).to(_lowerCAmelCase )
# forward pass
with torch.no_grad():
UpperCAmelCase_ =model(**_lowerCAmelCase )
UpperCAmelCase_ =outputs.logits
# verify the logits
UpperCAmelCase_ =torch.Size((1, 150, 160, 160) )
self.assertEqual(logits.shape , _lowerCAmelCase )
UpperCAmelCase_ =version.parse(PIL.__version__ ) < version.parse("9.0.0" )
if is_pillow_less_than_a:
UpperCAmelCase_ =torch.tensor(
[
[[-4.92_25, -2.39_54, -3.05_22], [-2.88_22, -1.00_46, -1.75_61], [-2.95_49, -1.32_28, -2.13_47]],
[[-5.81_68, -3.41_29, -4.07_78], [-3.86_51, -2.22_14, -3.02_77], [-3.83_56, -2.46_43, -3.35_35]],
[[-0.00_78, 3.99_52, 4.07_54], [2.98_56, 4.69_44, 5.00_35], [3.24_13, 4.78_13, 4.99_69]],
] , device=_lowerCAmelCase , )
else:
UpperCAmelCase_ =torch.tensor(
[
[[-4.89_60, -2.36_88, -3.03_55], [-2.84_78, -0.98_36, -1.74_18], [-2.94_49, -1.33_32, -2.14_56]],
[[-5.80_81, -3.41_24, -4.10_06], [-3.85_61, -2.20_81, -3.03_23], [-3.83_65, -2.46_01, -3.36_69]],
[[-0.03_09, 3.98_68, 4.05_40], [2.96_40, 4.68_77, 4.99_76], [3.20_81, 4.76_90, 4.99_42]],
] , device=_lowerCAmelCase , )
self.assertTrue(torch.allclose(logits[0, :3, :3, :3] , _lowerCAmelCase , atol=1e-4 ) )
@slow
def lowerCAmelCase__ ( self: int ) -> Union[str, Any]:
'''simple docstring'''
UpperCAmelCase_ =BeitForSemanticSegmentation.from_pretrained("microsoft/beit-base-finetuned-ade-640-640" )
UpperCAmelCase_ =model.to(_lowerCAmelCase )
UpperCAmelCase_ =BeitImageProcessor(do_resize=_lowerCAmelCase , size=640 , do_center_crop=_lowerCAmelCase )
UpperCAmelCase_ =load_dataset("hf-internal-testing/fixtures_ade20k" , split="test" )
UpperCAmelCase_ =Image.open(ds[0]["file"] )
UpperCAmelCase_ =image_processor(images=_lowerCAmelCase , return_tensors="pt" ).to(_lowerCAmelCase )
# forward pass
with torch.no_grad():
UpperCAmelCase_ =model(**_lowerCAmelCase )
UpperCAmelCase_ =outputs.logits.detach().cpu()
UpperCAmelCase_ =image_processor.post_process_semantic_segmentation(outputs=_lowerCAmelCase , target_sizes=[(500, 300)] )
UpperCAmelCase_ =torch.Size((500, 300) )
self.assertEqual(segmentation[0].shape , _lowerCAmelCase )
UpperCAmelCase_ =image_processor.post_process_semantic_segmentation(outputs=_lowerCAmelCase )
UpperCAmelCase_ =torch.Size((160, 160) )
self.assertEqual(segmentation[0].shape , _lowerCAmelCase )
| 54 |
import argparse
import random
import joblib
import numpy as np
import torch
from igf.igf import (
SecondaryLearner,
collect_objective_set,
compute_perplexity,
generate_datasets,
load_gpta,
recopy_gpta,
set_seed,
train_secondary_learner,
)
from torch.utils.data import DataLoader, RandomSampler
from transformers import GPTaLMHeadModel
def _snake_case (__lowercase=32 , __lowercase=10 , __lowercase=100 , __lowercase=1026 , __lowercase=True , __lowercase="data/tokenized_stories_train_wikitext103.jbl" , __lowercase="igf_context_pairs.jbl" , ):
set_seed(3)
# generate train_data and objective_set
UpperCamelCase_ , UpperCamelCase_ = generate_datasets(
__lowercase , __lowercase , number=__lowercase , min_len=1026 , trim=__lowercase)
# keeps model same across runs
set_seed(4)
# model, lm_optimizer, lm_scheduler = recopy_gpt2(model, device, max_steps) # store original model weights
# can we train on GPU?
UpperCamelCase_ = torch.device('cuda:0' if torch.cuda.is_available() else 'cpu')
# load pretrained model
UpperCamelCase_ = load_gpta('gpt2').to(__lowercase)
print('computing perplexity on objective set')
UpperCamelCase_ = compute_perplexity(__lowercase , __lowercase , __lowercase).item()
print('perplexity on objective set:' , __lowercase)
# collect igf pairs and save to file demo.jbl
collect_objective_set(__lowercase , __lowercase , __lowercase , __lowercase , __lowercase , __lowercase , __lowercase , __lowercase)
# clean up, delete model and data we don't need anymore
del model, train_data, objective_set
torch.cuda.empty_cache()
def _snake_case (__lowercase , __lowercase=15 , __lowercase=128 , __lowercase=100 , __lowercase="igf_model.pt" , ):
set_seed(42)
# Load pre-trained model
UpperCamelCase_ = GPTaLMHeadModel.from_pretrained('gpt2')
# Initialize secondary learner to use embedding weights of model
UpperCamelCase_ = SecondaryLearner(__lowercase)
# Train secondary learner
UpperCamelCase_ = train_secondary_learner(
__lowercase , __lowercase , max_epochs=__lowercase , batch_size=__lowercase , eval_freq=100 , igf_model_path=__lowercase , )
del model, secondary_learner_train_data
torch.cuda.empty_cache()
return secondary_learner
def _snake_case (__lowercase , __lowercase , __lowercase , __lowercase=32 , __lowercase=1000 , __lowercase=16 , __lowercase=1.0 , __lowercase=recopy_gpta , __lowercase=None , __lowercase=10 , __lowercase="gpt2_finetuned.pt" , ):
UpperCamelCase_ = torch.device('cuda:0' if torch.cuda.is_available() else 'cpu')
UpperCamelCase_ = RandomSampler(__lowercase)
UpperCamelCase_ = DataLoader(__lowercase , sampler=__lowercase)
UpperCamelCase_ = max_steps // (len(__lowercase)) + 1
UpperCamelCase_ = 0
UpperCamelCase_ = torch.zeros((1, context_len) , dtype=torch.long , device=__lowercase)
UpperCamelCase_ , UpperCamelCase_ , UpperCamelCase_ = recopy_model(__lowercase , __lowercase , __lowercase)
model.train()
if secondary_learner is not None:
secondary_learner.to(__lowercase)
secondary_learner.eval()
UpperCamelCase_ = []
UpperCamelCase_ = 0
UpperCamelCase_ = []
UpperCamelCase_ = []
# Compute the performance of the transformer model at the beginning
UpperCamelCase_ = compute_perplexity(__lowercase , __lowercase , __lowercase)
test_perps.append(__lowercase)
print('Test perplexity, step' , __lowercase , ':' , __lowercase)
for epoch in range(int(__lowercase)):
for step, example in enumerate(__lowercase):
torch.cuda.empty_cache()
UpperCamelCase_ = random.randint(0 , example.size(2) - context_len - 1)
UpperCamelCase_ = example[0, 0, start : start + context_len]
lm_optimizer.zero_grad()
UpperCamelCase_ = model(__lowercase , labels=__lowercase)
UpperCamelCase_ = True
if secondary_learner is not None:
UpperCamelCase_ = secondary_learner.forward(
torch.tensor(__lowercase , dtype=torch.long , device=__lowercase).unsqueeze(0))[0].item()
observed_qs.append(float(__lowercase))
# Here we implement the simple non-constant threshold for the predicted IG(X) value
# We will decay the selectivity of our secondary learner filter from
# 1 standard deviation above average to 1 below average after 10 batches.
if global_step == 10:
UpperCamelCase_ = -1
if predicted_q < threshold:
UpperCamelCase_ = False
# If we passed the filter, add the context to the batch!
if do_backprop:
contexts.append(np.array(context.cpu()))
UpperCamelCase_ = outputs[0]
lm_loss.backward()
examples += 1
del outputs
# Once the batch is filled with enough contexts, backprop on the batch.
if examples == batch_size:
torch.cuda.empty_cache()
UpperCamelCase_ = 0
# Do LM backprop
torch.nn.utils.clip_grad_norm_(model.parameters() , 3.0)
lm_optimizer.step()
lm_scheduler.step() # Update learning rate schedule
global_step += 1
# Compute the performance of the transformer model at this batch
if global_step % eval_interval == 0:
UpperCamelCase_ = compute_perplexity(__lowercase , __lowercase , __lowercase)
test_perps.append(__lowercase)
print('Test perplexity, step' , __lowercase , ':' , __lowercase)
# Break out of the loop after 60 batches
if max_steps > 0 and global_step > 60:
break
if max_steps > 0 and global_step > 60:
break
# save finetuned transformer model
torch.save(model.state_dict() , __lowercase)
torch.cuda.empty_cache()
# Do some cleaning up so we can reinitialize for the next run of this function
del lm_optimizer
del lm_scheduler
return model
def _snake_case ():
UpperCamelCase_ = argparse.ArgumentParser(description='Fine-tune a transformer model with IGF on a language modeling task')
# Required parameters
parser.add_argument(
'--data_dir' , default=__lowercase , type=__lowercase , required=__lowercase , help='The input data dir. Should contain data files for WikiText.' , )
parser.add_argument(
'--model_name_or_path' , default=__lowercase , type=__lowercase , required=__lowercase , help='Path to pretrained model or model identifier from huggingface.co/models' , )
parser.add_argument(
'--data_file' , type=__lowercase , default=__lowercase , help=(
'A jbl file containing tokenized data which can be split as objective dataset, '
'train_dataset and test_dataset.'
) , )
parser.add_argument(
'--igf_data_file' , type=__lowercase , default=__lowercase , help='A jbl file containing the context and information gain pairs to train secondary learner.' , )
parser.add_argument(
'--output_dir' , default=__lowercase , type=__lowercase , required=__lowercase , help='The output directory where the final fine-tuned model is stored.' , )
parser.add_argument(
'--tokenizer_name' , default=__lowercase , type=__lowercase , help='Pretrained tokenizer name or path if not the same as model_name' , )
parser.add_argument('--seed' , type=__lowercase , default=__lowercase , help='A seed for reproducible training.')
parser.add_argument(
'--context_len' , default=32 , type=__lowercase , help=(
'The maximum total input sequence length after tokenization. Sequences longer '
'than this will be truncated, sequences shorter will be padded.'
) , )
parser.add_argument(
'--size_objective_set' , default=100 , type=__lowercase , help='number of articles that are long enough to be used as our objective set' , )
parser.add_argument(
'--eval_freq' , default=100 , type=__lowercase , help='secondary model evaluation is triggered at eval_freq')
parser.add_argument('--max_steps' , default=1000 , type=__lowercase , help='To calculate training epochs')
parser.add_argument(
'--secondary_learner_batch_size' , default=128 , type=__lowercase , help='batch size of training data for secondary learner' , )
parser.add_argument(
'--batch_size' , default=16 , type=__lowercase , help='batch size of training data of language model(gpt2) ')
parser.add_argument(
'--eval_interval' , default=10 , type=__lowercase , help=(
'decay the selectivity of our secondary learner filter from'
'1 standard deviation above average to 1 below average after 10 batches'
) , )
parser.add_argument(
'--number' , default=100 , type=__lowercase , help='The number of examples split to be used as objective_set/test_data')
parser.add_argument(
'--min_len' , default=1026 , type=__lowercase , help='The minimum length of the article to be used as objective set')
parser.add_argument(
'--secondary_learner_max_epochs' , default=15 , type=__lowercase , help='number of epochs to train secondary learner')
parser.add_argument('--trim' , default=__lowercase , type=__lowercase , help='truncate the example if it exceeds context length')
parser.add_argument(
'--threshold' , default=1.0 , type=__lowercase , help=(
'The threshold value used by secondary learner to filter the train_data and allow only'
' informative data as input to the model'
) , )
parser.add_argument('--finetuned_model_name' , default='gpt2_finetuned.pt' , type=__lowercase , help='finetuned_model_name')
parser.add_argument(
'--recopy_model' , default=__lowercase , type=__lowercase , help='Reset the model to the original pretrained GPT-2 weights after each iteration' , )
# function calls
# Collecting *n* pairs of context and information gain(X, IG(X)) for training the secondary learner
generate_n_pairs(
context_len=32 , max_steps=10 , size_objective_set=100 , min_len=1026 , trim=__lowercase , data_file='data/tokenized_stories_train_wikitext103.jbl' , igf_data_file='igf_context_pairs.jbl' , )
# Load train data for secondary learner
UpperCamelCase_ = joblib.load('data/IGF_values.jbl')
# Train secondary learner
UpperCamelCase_ = training_secondary_learner(
__lowercase , secondary_learner_max_epochs=15 , secondary_learner_batch_size=128 , eval_freq=100 , igf_model_path='igf_model.pt' , )
# load pretrained gpt2 model
UpperCamelCase_ = GPTaLMHeadModel.from_pretrained('gpt2')
set_seed(42)
# Generate train and test data to train and evaluate gpt2 model
UpperCamelCase_ , UpperCamelCase_ = generate_datasets(
context_len=32 , file='data/tokenized_stories_train_wikitext103.jbl' , number=100 , min_len=1026 , trim=__lowercase)
# fine-tuning of the gpt2 model using igf (Information Gain Filtration)
finetune(
__lowercase , __lowercase , __lowercase , context_len=32 , max_steps=1000 , batch_size=16 , threshold=1.0 , recopy_model=__lowercase , secondary_learner=__lowercase , eval_interval=10 , finetuned_model_name='gpt2_finetuned.pt' , )
if __name__ == "__main__":
main()
| 23 | 0 |
def UpperCAmelCase ( a_ , a_ ) -> str:
"""simple docstring"""
if a < 0 or b < 0:
raise ValueError("the value of both inputs must be positive" )
__A = str(bin(a_ ) )[2:] # remove the leading "0b"
__A = str(bin(a_ ) )[2:] # remove the leading "0b"
__A = max(len(a_ ) , len(a_ ) )
return "0b" + "".join(
str(int(char_a == "1" and char_b == "1" ) )
for char_a, char_b in zip(a_binary.zfill(a_ ) , b_binary.zfill(a_ ) ) )
if __name__ == "__main__":
import doctest
doctest.testmod()
| 55 |
from __future__ import annotations
import unittest
from transformers import AutoTokenizer, MBartConfig, is_tf_available
from transformers.testing_utils import require_sentencepiece, require_tf, require_tokenizers, slow
from transformers.utils import cached_property
from ...test_configuration_common import ConfigTester
from ...test_modeling_tf_common import TFModelTesterMixin, ids_tensor
from ...test_pipeline_mixin import PipelineTesterMixin
if is_tf_available():
import tensorflow as tf
from transformers import TFAutoModelForSeqaSeqLM, TFMBartForConditionalGeneration, TFMBartModel
@require_tf
class _a :
"""simple docstring"""
A_ = MBartConfig
A_ = {}
A_ = """gelu"""
def __init__( self , _UpperCAmelCase , _UpperCAmelCase=13 , _UpperCAmelCase=7 , _UpperCAmelCase=True , _UpperCAmelCase=False , _UpperCAmelCase=99 , _UpperCAmelCase=32 , _UpperCAmelCase=2 , _UpperCAmelCase=4 , _UpperCAmelCase=37 , _UpperCAmelCase=0.1 , _UpperCAmelCase=0.1 , _UpperCAmelCase=20 , _UpperCAmelCase=2 , _UpperCAmelCase=1 , _UpperCAmelCase=0 , ) -> Union[str, Any]:
UpperCamelCase_ = parent
UpperCamelCase_ = batch_size
UpperCamelCase_ = seq_length
UpperCamelCase_ = is_training
UpperCamelCase_ = use_labels
UpperCamelCase_ = vocab_size
UpperCamelCase_ = hidden_size
UpperCamelCase_ = num_hidden_layers
UpperCamelCase_ = num_attention_heads
UpperCamelCase_ = intermediate_size
UpperCamelCase_ = hidden_dropout_prob
UpperCamelCase_ = attention_probs_dropout_prob
UpperCamelCase_ = max_position_embeddings
UpperCamelCase_ = eos_token_id
UpperCamelCase_ = pad_token_id
UpperCamelCase_ = bos_token_id
def _UpperCAmelCase ( self ) -> List[Any]:
UpperCamelCase_ = ids_tensor([self.batch_size, self.seq_length - 1] , self.vocab_size )
UpperCamelCase_ = tf.expand_dims(tf.constant([self.eos_token_id] * self.batch_size ) , 1 )
UpperCamelCase_ = tf.concat([input_ids, eos_tensor] , axis=1 )
UpperCamelCase_ = ids_tensor([self.batch_size, self.seq_length] , self.vocab_size )
UpperCamelCase_ = self.config_cls(
vocab_size=self.vocab_size , d_model=self.hidden_size , encoder_layers=self.num_hidden_layers , decoder_layers=self.num_hidden_layers , encoder_attention_heads=self.num_attention_heads , decoder_attention_heads=self.num_attention_heads , encoder_ffn_dim=self.intermediate_size , decoder_ffn_dim=self.intermediate_size , dropout=self.hidden_dropout_prob , attention_dropout=self.attention_probs_dropout_prob , max_position_embeddings=self.max_position_embeddings , eos_token_ids=[2] , bos_token_id=self.bos_token_id , pad_token_id=self.pad_token_id , decoder_start_token_id=self.pad_token_id , **self.config_updates , )
UpperCamelCase_ = prepare_mbart_inputs_dict(_UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase )
return config, inputs_dict
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> List[str]:
UpperCamelCase_ = TFMBartModel(config=_UpperCAmelCase ).get_decoder()
UpperCamelCase_ = inputs_dict['input_ids']
UpperCamelCase_ = input_ids[:1, :]
UpperCamelCase_ = inputs_dict['attention_mask'][:1, :]
UpperCamelCase_ = inputs_dict['head_mask']
UpperCamelCase_ = 1
# first forward pass
UpperCamelCase_ = model(_UpperCAmelCase , attention_mask=_UpperCAmelCase , head_mask=_UpperCAmelCase , use_cache=_UpperCAmelCase )
UpperCamelCase_ , UpperCamelCase_ = outputs.to_tuple()
UpperCamelCase_ = past_key_values[1]
def _snake_case (__lowercase , __lowercase , __lowercase , __lowercase=None , __lowercase=None , __lowercase=None , __lowercase=None , __lowercase=None , ):
if attention_mask is None:
UpperCamelCase_ = tf.cast(tf.math.not_equal(__lowercase , config.pad_token_id) , tf.inta)
if decoder_attention_mask is None:
UpperCamelCase_ = tf.concat(
[
tf.ones(decoder_input_ids[:, :1].shape , dtype=tf.inta),
tf.cast(tf.math.not_equal(decoder_input_ids[:, 1:] , config.pad_token_id) , tf.inta),
] , axis=-1 , )
if head_mask is None:
UpperCamelCase_ = tf.ones((config.encoder_layers, config.encoder_attention_heads))
if decoder_head_mask is None:
UpperCamelCase_ = tf.ones((config.decoder_layers, config.decoder_attention_heads))
if cross_attn_head_mask is None:
UpperCamelCase_ = tf.ones((config.decoder_layers, config.decoder_attention_heads))
return {
"input_ids": input_ids,
"decoder_input_ids": decoder_input_ids,
"attention_mask": attention_mask,
"decoder_attention_mask": decoder_attention_mask,
"head_mask": head_mask,
"decoder_head_mask": decoder_head_mask,
"cross_attn_head_mask": cross_attn_head_mask,
}
@require_tf
class _a ( UpperCAmelCase__ , UpperCAmelCase__ , unittest.TestCase ):
"""simple docstring"""
A_ = (TFMBartForConditionalGeneration, TFMBartModel) if is_tf_available() else ()
A_ = (TFMBartForConditionalGeneration,) if is_tf_available() else ()
A_ = (
{
"""conversational""": TFMBartForConditionalGeneration,
"""feature-extraction""": TFMBartModel,
"""summarization""": TFMBartForConditionalGeneration,
"""text2text-generation""": TFMBartForConditionalGeneration,
"""translation""": TFMBartForConditionalGeneration,
}
if is_tf_available()
else {}
)
A_ = True
A_ = False
A_ = False
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase ) -> Tuple:
if pipeline_test_casse_name != "FeatureExtractionPipelineTests":
# Exception encountered when calling layer '...'
return True
return False
def _UpperCAmelCase ( self ) -> Optional[int]:
UpperCamelCase_ = TFMBartModelTester(self )
UpperCamelCase_ = ConfigTester(self , config_class=_UpperCAmelCase )
def _UpperCAmelCase ( self ) -> Optional[int]:
self.config_tester.run_common_tests()
def _UpperCAmelCase ( self ) -> List[Any]:
UpperCamelCase_ = self.model_tester.prepare_config_and_inputs_for_common()
self.model_tester.check_decoder_model_past_large_inputs(*_UpperCAmelCase )
@require_sentencepiece
@require_tokenizers
@require_tf
class _a ( unittest.TestCase ):
"""simple docstring"""
A_ = [
""" UN Chief Says There Is No Military Solution in Syria""",
]
A_ = [
"""Şeful ONU declară că nu există o soluţie militară în Siria""",
]
A_ = """facebook/mbart-large-en-ro"""
@cached_property
def _UpperCAmelCase ( self ) -> Any:
return AutoTokenizer.from_pretrained(self.model_name )
@cached_property
def _UpperCAmelCase ( self ) -> List[str]:
UpperCamelCase_ = TFAutoModelForSeqaSeqLM.from_pretrained(self.model_name )
return model
def _UpperCAmelCase ( self , **_UpperCAmelCase ) -> int:
UpperCamelCase_ = self.translate_src_text(**_UpperCAmelCase )
self.assertListEqual(self.expected_text , _UpperCAmelCase )
def _UpperCAmelCase ( self , **_UpperCAmelCase ) -> List[str]:
UpperCamelCase_ = self.tokenizer(self.src_text , **_UpperCAmelCase , return_tensors='tf' )
UpperCamelCase_ = self.model.generate(
model_inputs.input_ids , attention_mask=model_inputs.attention_mask , num_beams=2 )
UpperCamelCase_ = self.tokenizer.batch_decode(_UpperCAmelCase , skip_special_tokens=_UpperCAmelCase )
return generated_words
@slow
def _UpperCAmelCase ( self ) -> List[Any]:
self._assert_generated_batch_equal_expected()
| 23 | 0 |
'''simple docstring'''
from __future__ import annotations
from PIL import Image
# Define glider example
_a : Union[str, Any] = [
[0, 1, 0, 0, 0, 0, 0, 0],
[0, 0, 1, 0, 0, 0, 0, 0],
[1, 1, 1, 0, 0, 0, 0, 0],
[0, 0, 0, 0, 0, 0, 0, 0],
[0, 0, 0, 0, 0, 0, 0, 0],
[0, 0, 0, 0, 0, 0, 0, 0],
[0, 0, 0, 0, 0, 0, 0, 0],
[0, 0, 0, 0, 0, 0, 0, 0],
]
# Define blinker example
_a : int = [[0, 1, 0], [0, 1, 0], [0, 1, 0]]
def _a (lowercase__ : list[list[int]] ) -> list[list[int]]:
"""simple docstring"""
__snake_case = []
for i in range(len(lowercase__ ) ):
__snake_case = []
for j in range(len(cells[i] ) ):
# Get the number of live neighbours
__snake_case = 0
if i > 0 and j > 0:
neighbour_count += cells[i - 1][j - 1]
if i > 0:
neighbour_count += cells[i - 1][j]
if i > 0 and j < len(cells[i] ) - 1:
neighbour_count += cells[i - 1][j + 1]
if j > 0:
neighbour_count += cells[i][j - 1]
if j < len(cells[i] ) - 1:
neighbour_count += cells[i][j + 1]
if i < len(lowercase__ ) - 1 and j > 0:
neighbour_count += cells[i + 1][j - 1]
if i < len(lowercase__ ) - 1:
neighbour_count += cells[i + 1][j]
if i < len(lowercase__ ) - 1 and j < len(cells[i] ) - 1:
neighbour_count += cells[i + 1][j + 1]
# Rules of the game of life (excerpt from Wikipedia):
# 1. Any live cell with two or three live neighbours survives.
# 2. Any dead cell with three live neighbours becomes a live cell.
# 3. All other live cells die in the next generation.
# Similarly, all other dead cells stay dead.
__snake_case = cells[i][j] == 1
if (
(alive and 2 <= neighbour_count <= 3)
or not alive
and neighbour_count == 3
):
next_generation_row.append(1 )
else:
next_generation_row.append(0 )
next_generation.append(lowercase__ )
return next_generation
def _a (lowercase__ : list[list[int]] , lowercase__ : int ) -> list[Image.Image]:
"""simple docstring"""
__snake_case = []
for _ in range(lowercase__ ):
# Create output image
__snake_case = Image.new('RGB' , (len(cells[0] ), len(lowercase__ )) )
__snake_case = img.load()
# Save cells to image
for x in range(len(lowercase__ ) ):
for y in range(len(cells[0] ) ):
__snake_case = 2_5_5 - cells[y][x] * 2_5_5
__snake_case = (colour, colour, colour)
# Save image
images.append(lowercase__ )
__snake_case = new_generation(lowercase__ )
return images
if __name__ == "__main__":
_a : Union[str, Any] = generate_images(GLIDER, 16)
images[0].save("out.gif", save_all=True, append_images=images[1:])
| 56 |
def _snake_case (__lowercase):
UpperCamelCase_ = 1
for i in range(1 , num + 1):
fact *= i
return fact
def _snake_case (__lowercase):
UpperCamelCase_ = 0
while number > 0:
UpperCamelCase_ = number % 10
sum_of_digits += last_digit
UpperCamelCase_ = number // 10 # Removing the last_digit from the given number
return sum_of_digits
def _snake_case (__lowercase = 100):
UpperCamelCase_ = factorial(__lowercase)
UpperCamelCase_ = split_and_add(__lowercase)
return result
if __name__ == "__main__":
print(solution(int(input("""Enter the Number: """).strip())))
| 23 | 0 |
import os
import shutil
import tempfile
import unittest
import numpy as np
from transformers import AutoTokenizer, BarkProcessor
from transformers.testing_utils import require_torch, slow
@require_torch
class _lowerCAmelCase( unittest.TestCase ):
"""simple docstring"""
def _a ( self ):
UpperCamelCase_: Union[str, Any] = 'ylacombe/bark-small'
UpperCamelCase_: Optional[int] = tempfile.mkdtemp()
UpperCamelCase_: Dict = 'en_speaker_1'
UpperCamelCase_: List[Any] = 'This is a test string'
UpperCamelCase_: Tuple = 'speaker_embeddings_path.json'
UpperCamelCase_: Tuple = 'speaker_embeddings'
def _a ( self , **_lowerCamelCase ):
return AutoTokenizer.from_pretrained(self.checkpoint , **_lowerCamelCase )
def _a ( self ):
shutil.rmtree(self.tmpdirname )
def _a ( self ):
UpperCamelCase_: int = self.get_tokenizer()
UpperCamelCase_: Dict = BarkProcessor(tokenizer=_lowerCamelCase )
processor.save_pretrained(self.tmpdirname )
UpperCamelCase_: Optional[Any] = BarkProcessor.from_pretrained(self.tmpdirname )
self.assertEqual(processor.tokenizer.get_vocab() , tokenizer.get_vocab() )
@slow
def _a ( self ):
UpperCamelCase_: Dict = BarkProcessor.from_pretrained(
pretrained_processor_name_or_path=self.checkpoint , speaker_embeddings_dict_path=self.speaker_embeddings_dict_path , )
processor.save_pretrained(
self.tmpdirname , speaker_embeddings_dict_path=self.speaker_embeddings_dict_path , speaker_embeddings_directory=self.speaker_embeddings_directory , )
UpperCamelCase_: Optional[int] = self.get_tokenizer(bos_token='(BOS)' , eos_token='(EOS)' )
UpperCamelCase_: Optional[int] = BarkProcessor.from_pretrained(
self.tmpdirname , self.speaker_embeddings_dict_path , bos_token='(BOS)' , eos_token='(EOS)' , )
self.assertEqual(processor.tokenizer.get_vocab() , tokenizer_add_kwargs.get_vocab() )
def _a ( self ):
UpperCamelCase_: Union[str, Any] = BarkProcessor.from_pretrained(
pretrained_processor_name_or_path=self.checkpoint , speaker_embeddings_dict_path=self.speaker_embeddings_dict_path , )
UpperCamelCase_: int = 3_5
UpperCamelCase_: Optional[int] = 2
UpperCamelCase_: int = 8
UpperCamelCase_: Union[str, Any] = {
'semantic_prompt': np.ones(_lowerCamelCase ),
'coarse_prompt': np.ones((nb_codebooks_coarse, seq_len) ),
'fine_prompt': np.ones((nb_codebooks_total, seq_len) ),
}
# test providing already loaded voice_preset
UpperCamelCase_: Dict = processor(text=self.input_string , voice_preset=_lowerCamelCase )
UpperCamelCase_: str = inputs['history_prompt']
for key in voice_preset:
self.assertListEqual(voice_preset[key].tolist() , processed_voice_preset.get(_lowerCamelCase , np.array([] ) ).tolist() )
# test loading voice preset from npz file
UpperCamelCase_: Tuple = os.path.join(self.tmpdirname , 'file.npz' )
np.savez(_lowerCamelCase , **_lowerCamelCase )
UpperCamelCase_: List[Any] = processor(text=self.input_string , voice_preset=_lowerCamelCase )
UpperCamelCase_: Tuple = inputs['history_prompt']
for key in voice_preset:
self.assertListEqual(voice_preset[key].tolist() , processed_voice_preset.get(_lowerCamelCase , np.array([] ) ).tolist() )
# test loading voice preset from the hub
UpperCamelCase_: int = processor(text=self.input_string , voice_preset=self.voice_preset )
def _a ( self ):
UpperCamelCase_: Tuple = self.get_tokenizer()
UpperCamelCase_: str = BarkProcessor(tokenizer=_lowerCamelCase )
UpperCamelCase_: List[str] = processor(text=self.input_string )
UpperCamelCase_: List[Any] = tokenizer(
self.input_string , padding='max_length' , max_length=2_5_6 , add_special_tokens=_lowerCamelCase , return_attention_mask=_lowerCamelCase , return_token_type_ids=_lowerCamelCase , )
for key in encoded_tok.keys():
self.assertListEqual(encoded_tok[key] , encoded_processor[key].squeeze().tolist() ) | 57 |
from typing import Dict, List, Optional, Union
import numpy as np
from ...image_processing_utils import BaseImageProcessor, BatchFeature, get_size_dict
from ...image_transforms import (
center_crop,
get_resize_output_image_size,
normalize,
rescale,
resize,
to_channel_dimension_format,
)
from ...image_utils import (
IMAGENET_STANDARD_MEAN,
IMAGENET_STANDARD_STD,
ChannelDimension,
ImageInput,
PILImageResampling,
is_valid_image,
to_numpy_array,
valid_images,
)
from ...utils import TensorType, is_vision_available, logging
if is_vision_available():
import PIL
snake_case__ : str = logging.get_logger(__name__)
def _snake_case (__lowercase):
if isinstance(__lowercase , (list, tuple)) and isinstance(videos[0] , (list, tuple)) and is_valid_image(videos[0][0]):
return videos
elif isinstance(__lowercase , (list, tuple)) and is_valid_image(videos[0]):
return [videos]
elif is_valid_image(__lowercase):
return [[videos]]
raise ValueError(f"""Could not make batched video from {videos}""")
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
A_ = ["""pixel_values"""]
def __init__( self , _UpperCAmelCase = True , _UpperCAmelCase = None , _UpperCAmelCase = PILImageResampling.BILINEAR , _UpperCAmelCase = True , _UpperCAmelCase = None , _UpperCAmelCase = True , _UpperCAmelCase = 1 / 255 , _UpperCAmelCase = True , _UpperCAmelCase = None , _UpperCAmelCase = None , **_UpperCAmelCase , ) -> None:
super().__init__(**_UpperCAmelCase )
UpperCamelCase_ = size if size is not None else {'shortest_edge': 224}
UpperCamelCase_ = get_size_dict(_UpperCAmelCase , default_to_square=_UpperCAmelCase )
UpperCamelCase_ = crop_size if crop_size is not None else {'height': 224, 'width': 224}
UpperCamelCase_ = get_size_dict(_UpperCAmelCase , param_name='crop_size' )
UpperCamelCase_ = do_resize
UpperCamelCase_ = size
UpperCamelCase_ = do_center_crop
UpperCamelCase_ = crop_size
UpperCamelCase_ = resample
UpperCamelCase_ = do_rescale
UpperCamelCase_ = rescale_factor
UpperCamelCase_ = do_normalize
UpperCamelCase_ = image_mean if image_mean is not None else IMAGENET_STANDARD_MEAN
UpperCamelCase_ = image_std if image_std is not None else IMAGENET_STANDARD_STD
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase = PILImageResampling.BILINEAR , _UpperCAmelCase = None , **_UpperCAmelCase , ) -> np.ndarray:
UpperCamelCase_ = get_size_dict(_UpperCAmelCase , default_to_square=_UpperCAmelCase )
if "shortest_edge" in size:
UpperCamelCase_ = get_resize_output_image_size(_UpperCAmelCase , size['shortest_edge'] , default_to_square=_UpperCAmelCase )
elif "height" in size and "width" in size:
UpperCamelCase_ = (size['height'], size['width'])
else:
raise ValueError(f"""Size must have 'height' and 'width' or 'shortest_edge' as keys. Got {size.keys()}""" )
return resize(_UpperCAmelCase , size=_UpperCAmelCase , resample=_UpperCAmelCase , data_format=_UpperCAmelCase , **_UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase = None , **_UpperCAmelCase , ) -> np.ndarray:
UpperCamelCase_ = get_size_dict(_UpperCAmelCase )
if "height" not in size or "width" not in size:
raise ValueError(f"""Size must have 'height' and 'width' as keys. Got {size.keys()}""" )
return center_crop(_UpperCAmelCase , size=(size['height'], size['width']) , data_format=_UpperCAmelCase , **_UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase = None , **_UpperCAmelCase , ) -> int:
return rescale(_UpperCAmelCase , scale=_UpperCAmelCase , data_format=_UpperCAmelCase , **_UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase = None , **_UpperCAmelCase , ) -> np.ndarray:
return normalize(_UpperCAmelCase , mean=_UpperCAmelCase , std=_UpperCAmelCase , data_format=_UpperCAmelCase , **_UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = ChannelDimension.FIRST , ) -> np.ndarray:
if do_resize and size is None or resample is None:
raise ValueError('Size and resample must be specified if do_resize is True.' )
if do_center_crop and crop_size is None:
raise ValueError('Crop size must be specified if do_center_crop is True.' )
if do_rescale and rescale_factor is None:
raise ValueError('Rescale factor must be specified if do_rescale is True.' )
if do_normalize and (image_mean is None or image_std is None):
raise ValueError('Image mean and std must be specified if do_normalize is True.' )
# All transformations expect numpy arrays.
UpperCamelCase_ = to_numpy_array(_UpperCAmelCase )
if do_resize:
UpperCamelCase_ = self.resize(image=_UpperCAmelCase , size=_UpperCAmelCase , resample=_UpperCAmelCase )
if do_center_crop:
UpperCamelCase_ = self.center_crop(_UpperCAmelCase , size=_UpperCAmelCase )
if do_rescale:
UpperCamelCase_ = self.rescale(image=_UpperCAmelCase , scale=_UpperCAmelCase )
if do_normalize:
UpperCamelCase_ = self.normalize(image=_UpperCAmelCase , mean=_UpperCAmelCase , std=_UpperCAmelCase )
UpperCamelCase_ = to_channel_dimension_format(_UpperCAmelCase , _UpperCAmelCase )
return image
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = ChannelDimension.FIRST , **_UpperCAmelCase , ) -> PIL.Image.Image:
UpperCamelCase_ = do_resize if do_resize is not None else self.do_resize
UpperCamelCase_ = resample if resample is not None else self.resample
UpperCamelCase_ = do_center_crop if do_center_crop is not None else self.do_center_crop
UpperCamelCase_ = do_rescale if do_rescale is not None else self.do_rescale
UpperCamelCase_ = rescale_factor if rescale_factor is not None else self.rescale_factor
UpperCamelCase_ = do_normalize if do_normalize is not None else self.do_normalize
UpperCamelCase_ = image_mean if image_mean is not None else self.image_mean
UpperCamelCase_ = image_std if image_std is not None else self.image_std
UpperCamelCase_ = size if size is not None else self.size
UpperCamelCase_ = get_size_dict(_UpperCAmelCase , default_to_square=_UpperCAmelCase )
UpperCamelCase_ = crop_size if crop_size is not None else self.crop_size
UpperCamelCase_ = get_size_dict(_UpperCAmelCase , param_name='crop_size' )
if not valid_images(_UpperCAmelCase ):
raise ValueError(
'Invalid image type. Must be of type PIL.Image.Image, numpy.ndarray, '
'torch.Tensor, tf.Tensor or jax.ndarray.' )
UpperCamelCase_ = make_batched(_UpperCAmelCase )
UpperCamelCase_ = [
[
self._preprocess_image(
image=_UpperCAmelCase , do_resize=_UpperCAmelCase , size=_UpperCAmelCase , resample=_UpperCAmelCase , do_center_crop=_UpperCAmelCase , crop_size=_UpperCAmelCase , do_rescale=_UpperCAmelCase , rescale_factor=_UpperCAmelCase , do_normalize=_UpperCAmelCase , image_mean=_UpperCAmelCase , image_std=_UpperCAmelCase , data_format=_UpperCAmelCase , )
for img in video
]
for video in videos
]
UpperCamelCase_ = {'pixel_values': videos}
return BatchFeature(data=_UpperCAmelCase , tensor_type=_UpperCAmelCase )
| 23 | 0 |
"""simple docstring"""
from __future__ import annotations
__lowerCAmelCase : List[Any] = 10
def __lowerCAmelCase ( __UpperCamelCase : list[int] ):
'''simple docstring'''
snake_case_ : Optional[Any] = 1
snake_case_ : Any = max(__UpperCamelCase )
while placement <= max_digit:
# declare and initialize empty buckets
snake_case_ : list[list] = [[] for _ in range(__UpperCamelCase )]
# split list_of_ints between the buckets
for i in list_of_ints:
snake_case_ : str = int((i / placement) % RADIX )
buckets[tmp].append(__UpperCamelCase )
# put each buckets' contents into list_of_ints
snake_case_ : Optional[int] = 0
for b in range(__UpperCamelCase ):
for i in buckets[b]:
snake_case_ : str = i
a += 1
# move to next
placement *= RADIX
return list_of_ints
if __name__ == "__main__":
import doctest
doctest.testmod()
| 58 |
# DISCLAIMER: This file is strongly influenced by https://github.com/yang-song/score_sde_pytorch
import math
from dataclasses import dataclass
from typing import Optional, Tuple, Union
import torch
from ..configuration_utils import ConfigMixin, register_to_config
from ..utils import BaseOutput, randn_tensor
from .scheduling_utils import SchedulerMixin, SchedulerOutput
@dataclass
class _a ( UpperCAmelCase__ ):
"""simple docstring"""
A_ = 42
A_ = 42
class _a ( UpperCAmelCase__ , UpperCAmelCase__ ):
"""simple docstring"""
A_ = 1
@register_to_config
def __init__( self , _UpperCAmelCase = 2000 , _UpperCAmelCase = 0.1_5 , _UpperCAmelCase = 0.0_1 , _UpperCAmelCase = 1_3_4_8.0 , _UpperCAmelCase = 1e-5 , _UpperCAmelCase = 1 , ) -> Tuple:
# standard deviation of the initial noise distribution
UpperCamelCase_ = sigma_max
# setable values
UpperCamelCase_ = None
self.set_sigmas(_UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase = None ) -> torch.FloatTensor:
return sample
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase = None , _UpperCAmelCase = None ) -> str:
UpperCamelCase_ = sampling_eps if sampling_eps is not None else self.config.sampling_eps
UpperCamelCase_ = torch.linspace(1 , _UpperCAmelCase , _UpperCAmelCase , device=_UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase = None , _UpperCAmelCase = None , _UpperCAmelCase = None ) -> Any:
UpperCamelCase_ = sigma_min if sigma_min is not None else self.config.sigma_min
UpperCamelCase_ = sigma_max if sigma_max is not None else self.config.sigma_max
UpperCamelCase_ = sampling_eps if sampling_eps is not None else self.config.sampling_eps
if self.timesteps is None:
self.set_timesteps(_UpperCAmelCase , _UpperCAmelCase )
UpperCamelCase_ = sigma_min * (sigma_max / sigma_min) ** (self.timesteps / sampling_eps)
UpperCamelCase_ = torch.exp(torch.linspace(math.log(_UpperCAmelCase ) , math.log(_UpperCAmelCase ) , _UpperCAmelCase ) )
UpperCamelCase_ = torch.tensor([sigma_min * (sigma_max / sigma_min) ** t for t in self.timesteps] )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> List[Any]:
return torch.where(
timesteps == 0 , torch.zeros_like(t.to(timesteps.device ) ) , self.discrete_sigmas[timesteps - 1].to(timesteps.device ) , )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase = None , _UpperCAmelCase = True , ) -> Union[SdeVeOutput, Tuple]:
if self.timesteps is None:
raise ValueError(
'`self.timesteps` is not set, you need to run \'set_timesteps\' after creating the scheduler' )
UpperCamelCase_ = timestep * torch.ones(
sample.shape[0] , device=sample.device ) # torch.repeat_interleave(timestep, sample.shape[0])
UpperCamelCase_ = (timestep * (len(self.timesteps ) - 1)).long()
# mps requires indices to be in the same device, so we use cpu as is the default with cuda
UpperCamelCase_ = timesteps.to(self.discrete_sigmas.device )
UpperCamelCase_ = self.discrete_sigmas[timesteps].to(sample.device )
UpperCamelCase_ = self.get_adjacent_sigma(_UpperCAmelCase , _UpperCAmelCase ).to(sample.device )
UpperCamelCase_ = torch.zeros_like(_UpperCAmelCase )
UpperCamelCase_ = (sigma**2 - adjacent_sigma**2) ** 0.5
# equation 6 in the paper: the model_output modeled by the network is grad_x log pt(x)
# also equation 47 shows the analog from SDE models to ancestral sampling methods
UpperCamelCase_ = diffusion.flatten()
while len(diffusion.shape ) < len(sample.shape ):
UpperCamelCase_ = diffusion.unsqueeze(-1 )
UpperCamelCase_ = drift - diffusion**2 * model_output
# equation 6: sample noise for the diffusion term of
UpperCamelCase_ = randn_tensor(
sample.shape , layout=sample.layout , generator=_UpperCAmelCase , device=sample.device , dtype=sample.dtype )
UpperCamelCase_ = sample - drift # subtract because `dt` is a small negative timestep
# TODO is the variable diffusion the correct scaling term for the noise?
UpperCamelCase_ = prev_sample_mean + diffusion * noise # add impact of diffusion field g
if not return_dict:
return (prev_sample, prev_sample_mean)
return SdeVeOutput(prev_sample=_UpperCAmelCase , prev_sample_mean=_UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase = None , _UpperCAmelCase = True , ) -> Union[SchedulerOutput, Tuple]:
if self.timesteps is None:
raise ValueError(
'`self.timesteps` is not set, you need to run \'set_timesteps\' after creating the scheduler' )
# For small batch sizes, the paper "suggest replacing norm(z) with sqrt(d), where d is the dim. of z"
# sample noise for correction
UpperCamelCase_ = randn_tensor(sample.shape , layout=sample.layout , generator=_UpperCAmelCase ).to(sample.device )
# compute step size from the model_output, the noise, and the snr
UpperCamelCase_ = torch.norm(model_output.reshape(model_output.shape[0] , -1 ) , dim=-1 ).mean()
UpperCamelCase_ = torch.norm(noise.reshape(noise.shape[0] , -1 ) , dim=-1 ).mean()
UpperCamelCase_ = (self.config.snr * noise_norm / grad_norm) ** 2 * 2
UpperCamelCase_ = step_size * torch.ones(sample.shape[0] ).to(sample.device )
# self.repeat_scalar(step_size, sample.shape[0])
# compute corrected sample: model_output term and noise term
UpperCamelCase_ = step_size.flatten()
while len(step_size.shape ) < len(sample.shape ):
UpperCamelCase_ = step_size.unsqueeze(-1 )
UpperCamelCase_ = sample + step_size * model_output
UpperCamelCase_ = prev_sample_mean + ((step_size * 2) ** 0.5) * noise
if not return_dict:
return (prev_sample,)
return SchedulerOutput(prev_sample=_UpperCAmelCase )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , ) -> torch.FloatTensor:
# Make sure sigmas and timesteps have the same device and dtype as original_samples
UpperCamelCase_ = timesteps.to(original_samples.device )
UpperCamelCase_ = self.discrete_sigmas.to(original_samples.device )[timesteps]
UpperCamelCase_ = (
noise * sigmas[:, None, None, None]
if noise is not None
else torch.randn_like(_UpperCAmelCase ) * sigmas[:, None, None, None]
)
UpperCamelCase_ = noise + original_samples
return noisy_samples
def __len__( self ) -> Optional[int]:
return self.config.num_train_timesteps
| 23 | 0 |
from __future__ import annotations
def lowerCAmelCase_ ( __a ) -> bool:
"""simple docstring"""
lowerCamelCase__: Union[str, Any] =len(__a )
# We need to create solution object to save path.
lowerCamelCase__: Optional[Any] =[[0 for _ in range(__a )] for _ in range(__a )]
lowerCamelCase__: int =run_maze(__a , 0 , 0 , __a )
if solved:
print("\n".join(str(__a ) for row in solutions ) )
else:
print("No solution exists!" )
return solved
def lowerCAmelCase_ ( __a , __a , __a , __a ) -> bool:
"""simple docstring"""
lowerCamelCase__: List[str] =len(__a )
# Final check point.
if i == j == (size - 1):
lowerCamelCase__: List[str] =1
return True
lowerCamelCase__: Optional[int] =(not i < 0) and (not j < 0) # Check lower bounds
lowerCamelCase__: Dict =(i < size) and (j < size) # Check upper bounds
if lower_flag and upper_flag:
# check for already visited and block points.
lowerCamelCase__: Optional[Any] =(not solutions[i][j]) and (not maze[i][j])
if block_flag:
# check visited
lowerCamelCase__: Optional[Any] =1
# check for directions
if (
run_maze(__a , i + 1 , __a , __a )
or run_maze(__a , __a , j + 1 , __a )
or run_maze(__a , i - 1 , __a , __a )
or run_maze(__a , __a , j - 1 , __a )
):
return True
lowerCamelCase__: List[Any] =0
return False
return False
if __name__ == "__main__":
import doctest
doctest.testmod()
| 59 |
from typing import TYPE_CHECKING
from ...utils import OptionalDependencyNotAvailable, _LazyModule, is_torch_available
snake_case__ : Optional[int] = {
"""configuration_pegasus_x""": ["""PEGASUS_X_PRETRAINED_CONFIG_ARCHIVE_MAP""", """PegasusXConfig"""],
}
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
snake_case__ : Dict = [
"""PEGASUS_X_PRETRAINED_MODEL_ARCHIVE_LIST""",
"""PegasusXForConditionalGeneration""",
"""PegasusXModel""",
"""PegasusXPreTrainedModel""",
]
if TYPE_CHECKING:
from .configuration_pegasus_x import PEGASUS_X_PRETRAINED_CONFIG_ARCHIVE_MAP, PegasusXConfig
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_pegasus_x import (
PEGASUS_X_PRETRAINED_MODEL_ARCHIVE_LIST,
PegasusXForConditionalGeneration,
PegasusXModel,
PegasusXPreTrainedModel,
)
else:
import sys
snake_case__ : Optional[Any] = _LazyModule(__name__, globals()["""__file__"""], _import_structure, module_spec=__spec__)
| 23 | 0 |
import importlib
import json
import os
from collections import OrderedDict
from typing import Dict, Optional, Union
# Build the list of all feature extractors
from ...configuration_utils import PretrainedConfig
from ...dynamic_module_utils import get_class_from_dynamic_module, resolve_trust_remote_code
from ...feature_extraction_utils import FeatureExtractionMixin
from ...utils import CONFIG_NAME, FEATURE_EXTRACTOR_NAME, get_file_from_repo, logging
from .auto_factory import _LazyAutoMapping
from .configuration_auto import (
CONFIG_MAPPING_NAMES,
AutoConfig,
model_type_to_module_name,
replace_list_option_in_docstrings,
)
lowerCAmelCase_ = logging.get_logger(__name__)
lowerCAmelCase_ = OrderedDict(
[
('''audio-spectrogram-transformer''', '''ASTFeatureExtractor'''),
('''beit''', '''BeitFeatureExtractor'''),
('''chinese_clip''', '''ChineseCLIPFeatureExtractor'''),
('''clap''', '''ClapFeatureExtractor'''),
('''clip''', '''CLIPFeatureExtractor'''),
('''clipseg''', '''ViTFeatureExtractor'''),
('''conditional_detr''', '''ConditionalDetrFeatureExtractor'''),
('''convnext''', '''ConvNextFeatureExtractor'''),
('''cvt''', '''ConvNextFeatureExtractor'''),
('''data2vec-audio''', '''Wav2Vec2FeatureExtractor'''),
('''data2vec-vision''', '''BeitFeatureExtractor'''),
('''deformable_detr''', '''DeformableDetrFeatureExtractor'''),
('''deit''', '''DeiTFeatureExtractor'''),
('''detr''', '''DetrFeatureExtractor'''),
('''dinat''', '''ViTFeatureExtractor'''),
('''donut-swin''', '''DonutFeatureExtractor'''),
('''dpt''', '''DPTFeatureExtractor'''),
('''encodec''', '''EncodecFeatureExtractor'''),
('''flava''', '''FlavaFeatureExtractor'''),
('''glpn''', '''GLPNFeatureExtractor'''),
('''groupvit''', '''CLIPFeatureExtractor'''),
('''hubert''', '''Wav2Vec2FeatureExtractor'''),
('''imagegpt''', '''ImageGPTFeatureExtractor'''),
('''layoutlmv2''', '''LayoutLMv2FeatureExtractor'''),
('''layoutlmv3''', '''LayoutLMv3FeatureExtractor'''),
('''levit''', '''LevitFeatureExtractor'''),
('''maskformer''', '''MaskFormerFeatureExtractor'''),
('''mctct''', '''MCTCTFeatureExtractor'''),
('''mobilenet_v1''', '''MobileNetV1FeatureExtractor'''),
('''mobilenet_v2''', '''MobileNetV2FeatureExtractor'''),
('''mobilevit''', '''MobileViTFeatureExtractor'''),
('''nat''', '''ViTFeatureExtractor'''),
('''owlvit''', '''OwlViTFeatureExtractor'''),
('''perceiver''', '''PerceiverFeatureExtractor'''),
('''poolformer''', '''PoolFormerFeatureExtractor'''),
('''regnet''', '''ConvNextFeatureExtractor'''),
('''resnet''', '''ConvNextFeatureExtractor'''),
('''segformer''', '''SegformerFeatureExtractor'''),
('''sew''', '''Wav2Vec2FeatureExtractor'''),
('''sew-d''', '''Wav2Vec2FeatureExtractor'''),
('''speech_to_text''', '''Speech2TextFeatureExtractor'''),
('''speecht5''', '''SpeechT5FeatureExtractor'''),
('''swiftformer''', '''ViTFeatureExtractor'''),
('''swin''', '''ViTFeatureExtractor'''),
('''swinv2''', '''ViTFeatureExtractor'''),
('''table-transformer''', '''DetrFeatureExtractor'''),
('''timesformer''', '''VideoMAEFeatureExtractor'''),
('''tvlt''', '''TvltFeatureExtractor'''),
('''unispeech''', '''Wav2Vec2FeatureExtractor'''),
('''unispeech-sat''', '''Wav2Vec2FeatureExtractor'''),
('''van''', '''ConvNextFeatureExtractor'''),
('''videomae''', '''VideoMAEFeatureExtractor'''),
('''vilt''', '''ViltFeatureExtractor'''),
('''vit''', '''ViTFeatureExtractor'''),
('''vit_mae''', '''ViTFeatureExtractor'''),
('''vit_msn''', '''ViTFeatureExtractor'''),
('''wav2vec2''', '''Wav2Vec2FeatureExtractor'''),
('''wav2vec2-conformer''', '''Wav2Vec2FeatureExtractor'''),
('''wavlm''', '''Wav2Vec2FeatureExtractor'''),
('''whisper''', '''WhisperFeatureExtractor'''),
('''xclip''', '''CLIPFeatureExtractor'''),
('''yolos''', '''YolosFeatureExtractor'''),
]
)
lowerCAmelCase_ = _LazyAutoMapping(CONFIG_MAPPING_NAMES, FEATURE_EXTRACTOR_MAPPING_NAMES)
def lowerCamelCase_ ( _UpperCamelCase ) -> Any:
"""simple docstring"""
for module_name, extractors in FEATURE_EXTRACTOR_MAPPING_NAMES.items():
if class_name in extractors:
snake_case_ : Union[str, Any] = model_type_to_module_name(_UpperCamelCase )
snake_case_ : Dict = importlib.import_module(f'''.{module_name}''' , '''transformers.models''' )
try:
return getattr(_UpperCamelCase , _UpperCamelCase )
except AttributeError:
continue
for _, extractor in FEATURE_EXTRACTOR_MAPPING._extra_content.items():
if getattr(_UpperCamelCase , '''__name__''' , _UpperCamelCase ) == class_name:
return extractor
# We did not fine the class, but maybe it's because a dep is missing. In that case, the class will be in the main
# init and we return the proper dummy to get an appropriate error message.
snake_case_ : Union[str, Any] = importlib.import_module('''transformers''' )
if hasattr(_UpperCamelCase , _UpperCamelCase ):
return getattr(_UpperCamelCase , _UpperCamelCase )
return None
def lowerCamelCase_ ( _UpperCamelCase , _UpperCamelCase = None , _UpperCamelCase = False , _UpperCamelCase = False , _UpperCamelCase = None , _UpperCamelCase = None , _UpperCamelCase = None , _UpperCamelCase = False , **_UpperCamelCase , ) -> Optional[Any]:
"""simple docstring"""
snake_case_ : Optional[Any] = get_file_from_repo(
_UpperCamelCase , _UpperCamelCase , cache_dir=_UpperCamelCase , force_download=_UpperCamelCase , resume_download=_UpperCamelCase , proxies=_UpperCamelCase , use_auth_token=_UpperCamelCase , revision=_UpperCamelCase , local_files_only=_UpperCamelCase , )
if resolved_config_file is None:
logger.info(
'''Could not locate the feature extractor configuration file, will try to use the model config instead.''' )
return {}
with open(_UpperCamelCase , encoding='''utf-8''' ) as reader:
return json.load(_UpperCamelCase )
class __lowerCAmelCase :
def __init__(self ) -> Optional[Any]:
'''simple docstring'''
raise EnvironmentError(
'''AutoFeatureExtractor is designed to be instantiated '''
'''using the `AutoFeatureExtractor.from_pretrained(pretrained_model_name_or_path)` method.''' )
@classmethod
@replace_list_option_in_docstrings(__magic_name__ )
def lowerCamelCase (cls , __magic_name__ , **__magic_name__ ) -> Union[str, Any]:
'''simple docstring'''
snake_case_ : Dict = kwargs.pop('''config''' , __magic_name__ )
snake_case_ : Union[str, Any] = kwargs.pop('''trust_remote_code''' , __magic_name__ )
snake_case_ : Tuple = True
snake_case_ , snake_case_ : Tuple = FeatureExtractionMixin.get_feature_extractor_dict(__magic_name__ , **__magic_name__ )
snake_case_ : int = config_dict.get('''feature_extractor_type''' , __magic_name__ )
snake_case_ : str = None
if "AutoFeatureExtractor" in config_dict.get('''auto_map''' , {} ):
snake_case_ : Any = config_dict['''auto_map''']['''AutoFeatureExtractor''']
# If we don't find the feature extractor class in the feature extractor config, let's try the model config.
if feature_extractor_class is None and feature_extractor_auto_map is None:
if not isinstance(__magic_name__ , __magic_name__ ):
snake_case_ : Any = AutoConfig.from_pretrained(__magic_name__ , **__magic_name__ )
# It could be in `config.feature_extractor_type``
snake_case_ : List[Any] = getattr(__magic_name__ , '''feature_extractor_type''' , __magic_name__ )
if hasattr(__magic_name__ , '''auto_map''' ) and "AutoFeatureExtractor" in config.auto_map:
snake_case_ : List[str] = config.auto_map['''AutoFeatureExtractor''']
if feature_extractor_class is not None:
snake_case_ : List[Any] = feature_extractor_class_from_name(__magic_name__ )
snake_case_ : Optional[int] = feature_extractor_auto_map is not None
snake_case_ : Tuple = feature_extractor_class is not None or type(__magic_name__ ) in FEATURE_EXTRACTOR_MAPPING
snake_case_ : Optional[int] = resolve_trust_remote_code(
__magic_name__ , __magic_name__ , __magic_name__ , __magic_name__ )
if has_remote_code and trust_remote_code:
snake_case_ : int = get_class_from_dynamic_module(
__magic_name__ , __magic_name__ , **__magic_name__ )
snake_case_ : int = kwargs.pop('''code_revision''' , __magic_name__ )
if os.path.isdir(__magic_name__ ):
feature_extractor_class.register_for_auto_class()
return feature_extractor_class.from_dict(__magic_name__ , **__magic_name__ )
elif feature_extractor_class is not None:
return feature_extractor_class.from_dict(__magic_name__ , **__magic_name__ )
# Last try: we use the FEATURE_EXTRACTOR_MAPPING.
elif type(__magic_name__ ) in FEATURE_EXTRACTOR_MAPPING:
snake_case_ : Optional[Any] = FEATURE_EXTRACTOR_MAPPING[type(__magic_name__ )]
return feature_extractor_class.from_dict(__magic_name__ , **__magic_name__ )
raise ValueError(
F'''Unrecognized feature extractor in {pretrained_model_name_or_path}. Should have a '''
F'''`feature_extractor_type` key in its {FEATURE_EXTRACTOR_NAME} of {CONFIG_NAME}, or one of the following '''
F'''`model_type` keys in its {CONFIG_NAME}: {", ".join(c for c in FEATURE_EXTRACTOR_MAPPING_NAMES.keys() )}''' )
@staticmethod
def lowerCamelCase (__magic_name__ , __magic_name__ ) -> Union[str, Any]:
'''simple docstring'''
FEATURE_EXTRACTOR_MAPPING.register(__magic_name__ , __magic_name__ )
| 60 |
import datasets
from .evaluate import evaluate
snake_case__ : int = """\
@article{hendrycks2021cuad,
title={CUAD: An Expert-Annotated NLP Dataset for Legal Contract Review},
author={Dan Hendrycks and Collin Burns and Anya Chen and Spencer Ball},
journal={arXiv preprint arXiv:2103.06268},
year={2021}
}
"""
snake_case__ : Union[str, Any] = """
This metric wrap the official scoring script for version 1 of the Contract
Understanding Atticus Dataset (CUAD).
Contract Understanding Atticus Dataset (CUAD) v1 is a corpus of more than 13,000 labels in 510
commercial legal contracts that have been manually labeled to identify 41 categories of important
clauses that lawyers look for when reviewing contracts in connection with corporate transactions.
"""
snake_case__ : Any = """
Computes CUAD scores (EM, F1, AUPR, Precision@80%Recall, and Precision@90%Recall).
Args:
predictions: List of question-answers dictionaries with the following key-values:
- 'id': id of the question-answer pair as given in the references (see below)
- 'prediction_text': list of possible texts for the answer, as a list of strings
depending on a threshold on the confidence probability of each prediction.
references: List of question-answers dictionaries with the following key-values:
- 'id': id of the question-answer pair (see above),
- 'answers': a Dict in the CUAD dataset format
{
'text': list of possible texts for the answer, as a list of strings
'answer_start': list of start positions for the answer, as a list of ints
}
Note that answer_start values are not taken into account to compute the metric.
Returns:
'exact_match': Exact match (the normalized answer exactly match the gold answer)
'f1': The F-score of predicted tokens versus the gold answer
'aupr': Area Under the Precision-Recall curve
'prec_at_80_recall': Precision at 80% recall
'prec_at_90_recall': Precision at 90% recall
Examples:
>>> predictions = [{'prediction_text': ['The seller:', 'The buyer/End-User: Shenzhen LOHAS Supply Chain Management Co., Ltd.'], 'id': 'LohaCompanyltd_20191209_F-1_EX-10.16_11917878_EX-10.16_Supply Agreement__Parties'}]
>>> references = [{'answers': {'answer_start': [143, 49], 'text': ['The seller:', 'The buyer/End-User: Shenzhen LOHAS Supply Chain Management Co., Ltd.']}, 'id': 'LohaCompanyltd_20191209_F-1_EX-10.16_11917878_EX-10.16_Supply Agreement__Parties'}]
>>> cuad_metric = datasets.load_metric(\"cuad\")
>>> results = cuad_metric.compute(predictions=predictions, references=references)
>>> print(results)
{'exact_match': 100.0, 'f1': 100.0, 'aupr': 0.0, 'prec_at_80_recall': 1.0, 'prec_at_90_recall': 1.0}
"""
@datasets.utils.file_utils.add_start_docstrings(_DESCRIPTION , _KWARGS_DESCRIPTION )
class _a ( datasets.Metric ):
"""simple docstring"""
def _UpperCAmelCase ( self ) -> Optional[int]:
return datasets.MetricInfo(
description=_DESCRIPTION , citation=_CITATION , inputs_description=_KWARGS_DESCRIPTION , features=datasets.Features(
{
'predictions': {
'id': datasets.Value('string' ),
'prediction_text': datasets.features.Sequence(datasets.Value('string' ) ),
},
'references': {
'id': datasets.Value('string' ),
'answers': datasets.features.Sequence(
{
'text': datasets.Value('string' ),
'answer_start': datasets.Value('int32' ),
} ),
},
} ) , codebase_urls=['https://www.atticusprojectai.org/cuad'] , reference_urls=['https://www.atticusprojectai.org/cuad'] , )
def _UpperCAmelCase ( self , _UpperCAmelCase , _UpperCAmelCase ) -> Dict:
UpperCamelCase_ = {prediction['id']: prediction['prediction_text'] for prediction in predictions}
UpperCamelCase_ = [
{
'paragraphs': [
{
'qas': [
{
'answers': [{'text': answer_text} for answer_text in ref['answers']['text']],
'id': ref['id'],
}
for ref in references
]
}
]
}
]
UpperCamelCase_ = evaluate(dataset=_UpperCAmelCase , predictions=_UpperCAmelCase )
return score
| 23 | 0 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.