Spaces:
Runtime error
Runtime error
Update app.py
Browse files
app.py
CHANGED
@@ -11,9 +11,16 @@ import numpy as np
|
|
11 |
from PIL import Image
|
12 |
from transformers import AutoProcessor, VisionEncoderDecoderModel, Gemma3nForConditionalGeneration, pipeline
|
13 |
import torch
|
14 |
-
|
15 |
-
|
16 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
17 |
import os
|
18 |
import tempfile
|
19 |
import uuid
|
@@ -316,13 +323,20 @@ except Exception as e:
|
|
316 |
model_status = f"β Model failed to load: {str(e)}"
|
317 |
|
318 |
# Initialize embedding model for RAG
|
319 |
-
|
320 |
-
|
321 |
-
|
322 |
-
|
323 |
-
|
324 |
-
|
325 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
326 |
embedding_model = None
|
327 |
|
328 |
# Initialize chatbot model
|
@@ -369,7 +383,7 @@ embedding_model = None
|
|
369 |
# chatbot_model is initialized above
|
370 |
|
371 |
|
372 |
-
def chunk_document(text, chunk_size=
|
373 |
"""Split document into overlapping chunks for RAG"""
|
374 |
words = text.split()
|
375 |
chunks = []
|
@@ -387,8 +401,8 @@ def create_embeddings(chunks):
|
|
387 |
return None
|
388 |
|
389 |
try:
|
390 |
-
# Process in
|
391 |
-
batch_size =
|
392 |
embeddings = []
|
393 |
|
394 |
for i in range(0, len(chunks), batch_size):
|
@@ -401,10 +415,10 @@ def create_embeddings(chunks):
|
|
401 |
print(f"Error creating embeddings: {e}")
|
402 |
return None
|
403 |
|
404 |
-
def retrieve_relevant_chunks(question, chunks, embeddings, top_k=
|
405 |
"""Retrieve most relevant chunks for a question"""
|
406 |
if embedding_model is None or embeddings is None:
|
407 |
-
return chunks[:
|
408 |
|
409 |
try:
|
410 |
question_embedding = embedding_model.encode([question], show_progress_bar=False)
|
@@ -417,7 +431,7 @@ def retrieve_relevant_chunks(question, chunks, embeddings, top_k=2):
|
|
417 |
return relevant_chunks
|
418 |
except Exception as e:
|
419 |
print(f"Error retrieving chunks: {e}")
|
420 |
-
return chunks[:
|
421 |
|
422 |
def process_uploaded_pdf(pdf_file, progress=gr.Progress()):
|
423 |
"""Main processing function for uploaded PDF"""
|
@@ -467,10 +481,6 @@ def clear_all():
|
|
467 |
document_chunks = []
|
468 |
document_embeddings = None
|
469 |
|
470 |
-
# Clear GPU memory
|
471 |
-
if torch.cuda.is_available():
|
472 |
-
torch.cuda.empty_cache()
|
473 |
-
|
474 |
return None, "", gr.Tabs(visible=False)
|
475 |
|
476 |
|
@@ -676,23 +686,15 @@ with gr.Blocks(
|
|
676 |
input_len = inputs["input_ids"].shape[-1]
|
677 |
|
678 |
with torch.inference_mode():
|
679 |
-
# Clear cache before generation
|
680 |
-
if torch.cuda.is_available():
|
681 |
-
torch.cuda.empty_cache()
|
682 |
-
|
683 |
generation = chatbot_model.generate(
|
684 |
**inputs,
|
685 |
-
max_new_tokens=
|
686 |
do_sample=False,
|
687 |
temperature=0.7,
|
688 |
pad_token_id=chatbot_processor.tokenizer.pad_token_id,
|
689 |
-
use_cache=
|
690 |
)
|
691 |
generation = generation[0][input_len:]
|
692 |
-
|
693 |
-
# Clear cache after generation
|
694 |
-
if torch.cuda.is_available():
|
695 |
-
torch.cuda.empty_cache()
|
696 |
|
697 |
response = chatbot_processor.decode(generation, skip_special_tokens=True)
|
698 |
|
|
|
11 |
from PIL import Image
|
12 |
from transformers import AutoProcessor, VisionEncoderDecoderModel, Gemma3nForConditionalGeneration, pipeline
|
13 |
import torch
|
14 |
+
try:
|
15 |
+
from sentence_transformers import SentenceTransformer
|
16 |
+
import numpy as np
|
17 |
+
from sklearn.metrics.pairwise import cosine_similarity
|
18 |
+
RAG_DEPENDENCIES_AVAILABLE = True
|
19 |
+
except ImportError as e:
|
20 |
+
print(f"RAG dependencies not available: {e}")
|
21 |
+
print("Please install: pip install sentence-transformers scikit-learn")
|
22 |
+
RAG_DEPENDENCIES_AVAILABLE = False
|
23 |
+
SentenceTransformer = None
|
24 |
import os
|
25 |
import tempfile
|
26 |
import uuid
|
|
|
323 |
model_status = f"β Model failed to load: {str(e)}"
|
324 |
|
325 |
# Initialize embedding model for RAG
|
326 |
+
if RAG_DEPENDENCIES_AVAILABLE:
|
327 |
+
try:
|
328 |
+
print("Loading embedding model for RAG...")
|
329 |
+
# Use GPU for embedding model with 24GB VRAM
|
330 |
+
device = "cuda" if torch.cuda.is_available() else "cpu"
|
331 |
+
embedding_model = SentenceTransformer('all-MiniLM-L6-v2', device=device)
|
332 |
+
print(f"β
Embedding model loaded successfully ({device})")
|
333 |
+
except Exception as e:
|
334 |
+
print(f"β Error loading embedding model: {e}")
|
335 |
+
import traceback
|
336 |
+
traceback.print_exc()
|
337 |
+
embedding_model = None
|
338 |
+
else:
|
339 |
+
print("β RAG dependencies not available")
|
340 |
embedding_model = None
|
341 |
|
342 |
# Initialize chatbot model
|
|
|
383 |
# chatbot_model is initialized above
|
384 |
|
385 |
|
386 |
+
def chunk_document(text, chunk_size=500, overlap=50):
|
387 |
"""Split document into overlapping chunks for RAG"""
|
388 |
words = text.split()
|
389 |
chunks = []
|
|
|
401 |
return None
|
402 |
|
403 |
try:
|
404 |
+
# Process in larger batches with 24GB GPU
|
405 |
+
batch_size = 64
|
406 |
embeddings = []
|
407 |
|
408 |
for i in range(0, len(chunks), batch_size):
|
|
|
415 |
print(f"Error creating embeddings: {e}")
|
416 |
return None
|
417 |
|
418 |
+
def retrieve_relevant_chunks(question, chunks, embeddings, top_k=3):
|
419 |
"""Retrieve most relevant chunks for a question"""
|
420 |
if embedding_model is None or embeddings is None:
|
421 |
+
return chunks[:3] # Fallback to first 3 chunks
|
422 |
|
423 |
try:
|
424 |
question_embedding = embedding_model.encode([question], show_progress_bar=False)
|
|
|
431 |
return relevant_chunks
|
432 |
except Exception as e:
|
433 |
print(f"Error retrieving chunks: {e}")
|
434 |
+
return chunks[:3] # Fallback
|
435 |
|
436 |
def process_uploaded_pdf(pdf_file, progress=gr.Progress()):
|
437 |
"""Main processing function for uploaded PDF"""
|
|
|
481 |
document_chunks = []
|
482 |
document_embeddings = None
|
483 |
|
|
|
|
|
|
|
|
|
484 |
return None, "", gr.Tabs(visible=False)
|
485 |
|
486 |
|
|
|
686 |
input_len = inputs["input_ids"].shape[-1]
|
687 |
|
688 |
with torch.inference_mode():
|
|
|
|
|
|
|
|
|
689 |
generation = chatbot_model.generate(
|
690 |
**inputs,
|
691 |
+
max_new_tokens=400, # Increased for 24GB GPU
|
692 |
do_sample=False,
|
693 |
temperature=0.7,
|
694 |
pad_token_id=chatbot_processor.tokenizer.pad_token_id,
|
695 |
+
use_cache=True # Enable KV cache with more VRAM
|
696 |
)
|
697 |
generation = generation[0][input_len:]
|
|
|
|
|
|
|
|
|
698 |
|
699 |
response = chatbot_processor.decode(generation, skip_special_tokens=True)
|
700 |
|