Spaces:
Runtime error
Runtime error
File size: 3,867 Bytes
01e655b |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 |
import re
import torch
import torch.nn.functional as F
class CustomRepetitionPenaltyLogitsProcessorRepeat:
def __init__(self, penalty: float, max_input_ids, past_window):
if not isinstance(penalty, float) or not (penalty > 0):
raise ValueError(
f"`penalty` has to be a strictly positive float, but is {penalty}"
)
self.penalty = penalty
self.max_input_ids = max_input_ids
self.past_window = past_window
def __call__(
self, input_ids: torch.LongTensor, scores: torch.FloatTensor
) -> torch.FloatTensor:
input_ids = input_ids[:, -self.past_window :]
freq = F.one_hot(input_ids, scores.size(1)).sum(1)
freq[self.max_input_ids :] = 0
alpha = self.penalty**freq
scores = torch.where(scores < 0, scores * alpha, scores / alpha)
return scores
class CustomRepetitionPenaltyLogitsProcessor:
def __init__(self, penalty: float, max_input_ids, past_window):
if not isinstance(penalty, float) or not (penalty > 0):
raise ValueError(
f"`penalty` has to be a strictly positive float, but is {penalty}"
)
self.penalty = penalty
self.max_input_ids = max_input_ids
self.past_window = past_window
def __call__(
self, input_ids: torch.LongTensor, scores: torch.FloatTensor
) -> torch.FloatTensor:
input_ids = input_ids[:, -self.past_window :]
score = torch.gather(scores, 1, input_ids)
_score = score.detach().clone()
score = torch.where(score < 0, score * self.penalty, score / self.penalty)
score[input_ids >= self.max_input_ids] = _score[input_ids >= self.max_input_ids]
scores.scatter_(1, input_ids, score)
return scores
def count_invalid_characters(s, reserved_tokens: list = []):
escaped_tokens = [re.escape(token) for token in reserved_tokens]
reserved_pattern = "|".join(escaped_tokens)
s = re.sub(rf"{reserved_pattern}", "", s)
pattern = re.compile(r"[^\u4e00-\u9fffA-Za-z，。、,\. ]")
non_alphabetic_chinese_chars = pattern.findall(s)
return set(non_alphabetic_chinese_chars)
def detect_language(sentence):
chinese_char_pattern = re.compile(r"[\u4e00-\u9fff]")
english_word_pattern = re.compile(r"\b[A-Za-z]+\b")
chinese_chars = chinese_char_pattern.findall(sentence)
english_words = english_word_pattern.findall(sentence)
if len(chinese_chars) > len(english_words):
return "zh"
else:
return "en"
character_map = {
"：": "，",
"；": "，",
"！": "。",
"（": "，",
"）": "，",
"【": "，",
"】": "，",
"『": "，",
"』": "，",
"「": "，",
"」": "，",
"《": "，",
"》": "，",
"－": "，",
"‘": "",
"“": "",
"’": "",
"”": "",
":": ",",
";": ",",
"!": ".",
"(": ",",
")": ",",
# '[': ',',
# ']': ',',
">": ",",
"<": ",",
"-": ",",
}
halfwidth_2_fullwidth_map = {
"!": "！",
'"': "“",
"'": "‘",
"#": "＃",
"$": "＄",
"%": "％",
"&": "＆",
"(": "（",
")": "）",
",": "，",
"-": "－",
"*": "＊",
"+": "＋",
".": "。",
"/": "／",
":": "：",
";": "；",
"<": "＜",
"=": "＝",
">": "＞",
"?": "？",
"@": "＠",
# '[': '［',
"\\": "＼",
# ']': '］',
"^": "＾",
# '_': '＿',
"`": "｀",
"{": "｛",
"|": "｜",
"}": "｝",
"~": "～",
}
def apply_half2full_map(text):
translation_table = str.maketrans(halfwidth_2_fullwidth_map)
return text.translate(translation_table)
def apply_character_map(text):
translation_table = str.maketrans(character_map)
return text.translate(translation_table)
|