Spaces:
Runtime error
Runtime error
Update app.py
Browse files
app.py
CHANGED
@@ -1,6 +1,5 @@
|
|
1 |
import os
|
2 |
os.environ["TRANSFORMERS_NO_FAST"] = "1" # Force use of slow tokenizers
|
3 |
-
os.environ["CUDA_LAUNCH_BLOCKING"] = "1"
|
4 |
|
5 |
import io
|
6 |
import torch
|
@@ -14,7 +13,7 @@ import numpy as np
|
|
14 |
import json
|
15 |
import tempfile
|
16 |
from fastapi import FastAPI, UploadFile, File, HTTPException, Form, BackgroundTasks
|
17 |
-
from fastapi.responses import FileResponse, JSONResponse, HTMLResponse
|
18 |
from fastapi.middleware.cors import CORSMiddleware
|
19 |
from transformers import pipeline, AutoModelForQuestionAnswering, AutoTokenizer
|
20 |
from sentence_transformers import SentenceTransformer
|
@@ -28,17 +27,14 @@ import hashlib # For caching file results
|
|
28 |
# For asynchronous blocking calls
|
29 |
from starlette.concurrency import run_in_threadpool
|
30 |
|
31 |
-
#
|
32 |
import gensim
|
33 |
from gensim import corpora, models
|
34 |
|
35 |
-
# Spacy stop words
|
36 |
-
from spacy.lang.en.stop_words import STOP_WORDS
|
37 |
-
|
38 |
# Global cache for analysis results based on file hash
|
39 |
analysis_cache = {}
|
40 |
|
41 |
-
# Ensure compatibility with Google Colab
|
42 |
try:
|
43 |
from google.colab import drive
|
44 |
drive.mount('/content/drive')
|
@@ -49,7 +45,7 @@ except Exception:
|
|
49 |
os.makedirs("static", exist_ok=True)
|
50 |
os.makedirs("temp", exist_ok=True)
|
51 |
|
52 |
-
#
|
53 |
device = "cuda" if torch.cuda.is_available() else "cpu"
|
54 |
|
55 |
# Initialize FastAPI
|
@@ -68,13 +64,16 @@ app.add_middleware(
|
|
68 |
document_storage = {}
|
69 |
chat_history = []
|
70 |
|
|
|
71 |
def store_document_context(task_id, text):
|
72 |
document_storage[task_id] = text
|
73 |
return True
|
74 |
|
|
|
75 |
def load_document_context(task_id):
|
76 |
return document_storage.get(task_id, "")
|
77 |
|
|
|
78 |
def compute_md5(content: bytes) -> str:
|
79 |
return hashlib.md5(content).hexdigest()
|
80 |
|
@@ -84,6 +83,7 @@ def compute_md5(content: bytes) -> str:
|
|
84 |
|
85 |
def fine_tune_cuad_model():
|
86 |
from datasets import load_dataset
|
|
|
87 |
from transformers import Trainer, TrainingArguments, AutoModelForQuestionAnswering, AutoTokenizer
|
88 |
|
89 |
print("✅ Loading CUAD dataset for fine tuning...")
|
@@ -121,10 +121,7 @@ def fine_tune_cuad_model():
|
|
121 |
tokenized_examples["end_positions"] = []
|
122 |
for i, offsets in enumerate(offset_mapping):
|
123 |
input_ids = tokenized_examples["input_ids"][i]
|
124 |
-
|
125 |
-
cls_index = input_ids.index(tokenizer.cls_token_id)
|
126 |
-
except ValueError:
|
127 |
-
cls_index = 0
|
128 |
sequence_ids = tokenized_examples.sequence_ids(i)
|
129 |
sample_index = sample_mapping[i]
|
130 |
answers = examples["answers"][sample_index]
|
@@ -135,26 +132,21 @@ def fine_tune_cuad_model():
|
|
135 |
start_char = answers["answer_start"][0]
|
136 |
end_char = start_char + len(answers["text"][0])
|
137 |
tokenized_start_index = 0
|
138 |
-
while
|
139 |
tokenized_start_index += 1
|
140 |
tokenized_end_index = len(input_ids) - 1
|
141 |
-
while
|
142 |
tokenized_end_index -= 1
|
143 |
-
if
|
144 |
-
tokenized_examples["start_positions"].append(cls_index)
|
145 |
-
tokenized_examples["end_positions"].append(cls_index)
|
146 |
-
elif not (offsets[tokenized_start_index][0] <= start_char and offsets[tokenized_end_index][1] >= end_char):
|
147 |
tokenized_examples["start_positions"].append(cls_index)
|
148 |
tokenized_examples["end_positions"].append(cls_index)
|
149 |
else:
|
150 |
while tokenized_start_index < len(offsets) and offsets[tokenized_start_index][0] <= start_char:
|
151 |
tokenized_start_index += 1
|
152 |
-
|
153 |
-
|
154 |
-
while tokenized_end_index >= 0 and offsets[tokenized_end_index][1] >= end_char:
|
155 |
tokenized_end_index -= 1
|
156 |
-
|
157 |
-
tokenized_examples["end_positions"].append(safe_end)
|
158 |
return tokenized_examples
|
159 |
|
160 |
print("✅ Tokenizing dataset...")
|
@@ -198,70 +190,53 @@ def fine_tune_cuad_model():
|
|
198 |
#############################
|
199 |
|
200 |
try:
|
201 |
-
# Load spaCy model
|
202 |
try:
|
203 |
nlp = spacy.load("en_core_web_sm")
|
204 |
except Exception:
|
205 |
spacy.cli.download("en_core_web_sm")
|
206 |
nlp = spacy.load("en_core_web_sm")
|
207 |
-
print("✅
|
208 |
-
|
209 |
-
# Create summarizer and QA pipelines on GPU
|
210 |
summarizer = pipeline(
|
211 |
"summarization",
|
212 |
-
model="
|
213 |
-
tokenizer="
|
214 |
-
device=0 if
|
215 |
-
)
|
216 |
-
qa_model = pipeline(
|
217 |
-
"question-answering",
|
218 |
-
model="deepset/roberta-base-squad2",
|
219 |
-
device=0 if device == "cuda" else -1
|
220 |
)
|
|
|
|
|
|
|
221 |
|
222 |
-
# Use GPU for sentence embeddings if available
|
223 |
embedding_model = SentenceTransformer("all-mpnet-base-v2", device=device)
|
224 |
-
|
225 |
-
ner_model = pipeline("ner", model="dslim/bert-base-NER", device=0 if device == "cuda" else -1)
|
226 |
-
|
227 |
-
# Speech-to-text pipeline on GPU (if available)
|
228 |
speech_to_text = pipeline("automatic-speech-recognition", model="openai/whisper-medium", chunk_length_s=30,
|
229 |
-
device_map="auto" if
|
230 |
-
|
231 |
-
# Load or fine-tune the CUAD QA model and move to GPU
|
232 |
if os.path.exists("fine_tuned_legal_qa"):
|
233 |
print("✅ Loading fine-tuned CUAD QA model from fine_tuned_legal_qa...")
|
234 |
cuad_tokenizer = AutoTokenizer.from_pretrained("fine_tuned_legal_qa")
|
235 |
from transformers import AutoModelForQuestionAnswering
|
236 |
cuad_model = AutoModelForQuestionAnswering.from_pretrained("fine_tuned_legal_qa")
|
237 |
cuad_model.to(device)
|
|
|
|
|
238 |
else:
|
239 |
-
print("
|
240 |
cuad_tokenizer, cuad_model = fine_tune_cuad_model()
|
241 |
cuad_model.to(device)
|
242 |
-
|
243 |
-
sentiment_pipeline = pipeline(
|
244 |
-
"sentiment-analysis",
|
245 |
-
model="distilbert-base-uncased-finetuned-sst-2-english",
|
246 |
-
device=0 if device == "cuda" else -1
|
247 |
-
)
|
248 |
-
|
249 |
-
print("✅ All models loaded successfully.")
|
250 |
except Exception as e:
|
251 |
-
print(f"
|
252 |
raise RuntimeError(f"Error loading models: {str(e)}")
|
253 |
|
254 |
-
|
255 |
-
|
256 |
-
|
|
|
257 |
|
258 |
def legal_chatbot(user_input, context):
|
259 |
global chat_history
|
260 |
chat_history.append({"role": "user", "content": user_input})
|
261 |
-
|
262 |
-
response = qa_model(question=user_input, context=context)["answer"]
|
263 |
-
except Exception as e:
|
264 |
-
response = f"Error processing query: {e}"
|
265 |
chat_history.append({"role": "assistant", "content": response})
|
266 |
return response
|
267 |
|
@@ -314,9 +289,9 @@ def extract_named_entities(text):
|
|
314 |
entities.extend([{"entity": ent.text, "label": ent.label_} for ent in doc.ents])
|
315 |
return entities
|
316 |
|
317 |
-
|
318 |
-
#
|
319 |
-
|
320 |
|
321 |
def analyze_sentiment(text):
|
322 |
sentences = [sent.text for sent in nlp(text).sents]
|
@@ -343,82 +318,20 @@ def get_enhanced_context_info(text):
|
|
343 |
enhanced["topics"] = analyze_topics(text, num_topics=5)
|
344 |
return enhanced
|
345 |
|
346 |
-
def explain_topics(topics):
|
347 |
-
explanation = {}
|
348 |
-
for topic_idx, topic_str in topics:
|
349 |
-
parts = topic_str.split('+')
|
350 |
-
terms = []
|
351 |
-
for part in parts:
|
352 |
-
part = part.strip()
|
353 |
-
if '*' in part:
|
354 |
-
weight_str, word = part.split('*', 1)
|
355 |
-
word = word.strip().strip('\"').strip('\'')
|
356 |
-
try:
|
357 |
-
weight = float(weight_str)
|
358 |
-
except:
|
359 |
-
weight = 0.0
|
360 |
-
if word.lower() not in STOP_WORDS and len(word) > 1:
|
361 |
-
terms.append((weight, word))
|
362 |
-
terms.sort(key=lambda x: -x[0])
|
363 |
-
if terms:
|
364 |
-
if any("liability" in w.lower() for _, w in terms):
|
365 |
-
label = "Liability & Penalty Risk"
|
366 |
-
elif any("termination" in w.lower() for _, w in terms):
|
367 |
-
label = "Termination & Refund Risk"
|
368 |
-
elif any("compliance" in w.lower() for _, w in terms):
|
369 |
-
label = "Compliance & Regulatory Risk"
|
370 |
-
else:
|
371 |
-
label = "General Risk Language"
|
372 |
-
else:
|
373 |
-
label = "General Risk Language"
|
374 |
-
|
375 |
-
explanation_text = (
|
376 |
-
f"Topic {topic_idx} ({label}) is characterized by dominant terms: " +
|
377 |
-
", ".join([f"'{word}' ({weight:.3f})" for weight, word in terms[:5]])
|
378 |
-
)
|
379 |
-
explanation[topic_idx] = {
|
380 |
-
"label": label,
|
381 |
-
"explanation": explanation_text,
|
382 |
-
"terms": terms
|
383 |
-
}
|
384 |
-
return explanation
|
385 |
-
|
386 |
def analyze_risk_enhanced(text):
|
387 |
enhanced = get_enhanced_context_info(text)
|
388 |
avg_sentiment = enhanced["average_sentiment"]
|
389 |
risk_score = abs(avg_sentiment) if avg_sentiment < 0 else 0
|
390 |
-
|
391 |
-
topics_explanation = explain_topics(topics_raw)
|
392 |
-
return {
|
393 |
-
"risk_score": risk_score,
|
394 |
-
"average_sentiment": avg_sentiment,
|
395 |
-
"topics": topics_raw,
|
396 |
-
"topics_explanation": topics_explanation
|
397 |
-
}
|
398 |
|
399 |
-
|
400 |
-
#
|
401 |
-
|
402 |
-
|
403 |
-
def chunk_text_by_tokens(text, tokenizer, max_chunk_len=384, stride=128):
|
404 |
-
encoded = tokenizer(text, add_special_tokens=False)
|
405 |
-
input_ids = encoded["input_ids"]
|
406 |
-
chunks = []
|
407 |
-
idx = 0
|
408 |
-
while idx < len(input_ids):
|
409 |
-
end = idx + max_chunk_len
|
410 |
-
sub_ids = input_ids[idx:end]
|
411 |
-
chunk_text = tokenizer.decode(sub_ids, skip_special_tokens=True)
|
412 |
-
chunks.append(chunk_text)
|
413 |
-
if end >= len(input_ids):
|
414 |
-
break
|
415 |
-
idx = end - stride
|
416 |
-
if idx < 0:
|
417 |
-
idx = 0
|
418 |
-
return chunks
|
419 |
|
420 |
def analyze_contract_clauses(text):
|
421 |
-
|
|
|
|
|
422 |
try:
|
423 |
clause_types = list(cuad_model.config.id2label.values())
|
424 |
except Exception:
|
@@ -428,50 +341,26 @@ def analyze_contract_clauses(text):
|
|
428 |
"Assignment", "Warranty", "Limitation of Liability", "Arbitration",
|
429 |
"IP Rights", "Force Majeure", "Revenue/Profit Sharing", "Audit Rights"
|
430 |
]
|
431 |
-
|
432 |
-
|
433 |
-
for chunk in
|
434 |
-
chunk =
|
435 |
-
|
436 |
-
|
437 |
-
|
438 |
-
|
439 |
-
|
440 |
-
|
441 |
-
inputs["input_ids"] = torch.clamp(inputs["input_ids"], max=cuad_model.config.vocab_size - 1)
|
442 |
-
if torch.any(inputs["input_ids"] >= cuad_model.config.vocab_size):
|
443 |
-
print("Invalid token id found; skipping chunk")
|
444 |
-
continue
|
445 |
-
with torch.no_grad():
|
446 |
-
outputs = cuad_model(**inputs)
|
447 |
-
if device == "cuda":
|
448 |
-
torch.cuda.synchronize()
|
449 |
-
if outputs.start_logits.shape[1] != inputs["input_ids"].shape[1]:
|
450 |
-
print("Mismatch in logits shape; skipping chunk")
|
451 |
-
continue
|
452 |
-
predictions = torch.sigmoid(outputs.start_logits).cpu().numpy()[0]
|
453 |
-
for idx, confidence in enumerate(predictions):
|
454 |
-
if confidence > 0.5 and idx < len(clause_types):
|
455 |
-
clauses_detected.append({
|
456 |
-
"type": clause_types[idx],
|
457 |
-
"confidence": float(confidence)
|
458 |
-
})
|
459 |
-
except Exception as e:
|
460 |
-
print(f"Error processing chunk: {e}")
|
461 |
-
if device == "cuda":
|
462 |
-
torch.cuda.empty_cache()
|
463 |
-
continue
|
464 |
-
|
465 |
aggregated_clauses = {}
|
466 |
for clause in clauses_detected:
|
467 |
-
|
468 |
-
if
|
469 |
-
aggregated_clauses[
|
470 |
return list(aggregated_clauses.values())
|
471 |
|
472 |
-
|
473 |
-
#
|
474 |
-
|
475 |
|
476 |
@app.post("/analyze_legal_document")
|
477 |
async def analyze_legal_document(file: UploadFile = File(...)):
|
@@ -484,14 +373,7 @@ async def analyze_legal_document(file: UploadFile = File(...)):
|
|
484 |
if not text:
|
485 |
return {"status": "error", "message": "No valid text found in the document."}
|
486 |
summary_text = text[:4096] if len(text) > 4096 else text
|
487 |
-
|
488 |
-
if len(text) > 100:
|
489 |
-
summary = summarizer(summary_text, max_length=200, min_length=50, do_sample=False)[0]['summary_text']
|
490 |
-
else:
|
491 |
-
summary = "Document too short for a meaningful summary."
|
492 |
-
except Exception as e:
|
493 |
-
summary = "Summarization failed due to an error."
|
494 |
-
print(f"Summarization error: {e}")
|
495 |
entities = extract_named_entities(text)
|
496 |
risk_analysis = analyze_risk_enhanced(text)
|
497 |
clauses = analyze_contract_clauses(text)
|
@@ -529,14 +411,7 @@ async def analyze_legal_video(file: UploadFile = File(...), background_tasks: Ba
|
|
529 |
with open(transcript_path, "w") as f:
|
530 |
f.write(text)
|
531 |
summary_text = text[:4096] if len(text) > 4096 else text
|
532 |
-
|
533 |
-
if len(text) > 100:
|
534 |
-
summary = summarizer(summary_text, max_length=200, min_length=50, do_sample=False)[0]['summary_text']
|
535 |
-
else:
|
536 |
-
summary = "Transcript too short for meaningful summarization."
|
537 |
-
except Exception as e:
|
538 |
-
summary = "Summarization failed due to an error."
|
539 |
-
print(f"Summarization error: {e}")
|
540 |
entities = extract_named_entities(text)
|
541 |
risk_analysis = analyze_risk_enhanced(text)
|
542 |
clauses = analyze_contract_clauses(text)
|
@@ -576,14 +451,7 @@ async def analyze_legal_audio(file: UploadFile = File(...), background_tasks: Ba
|
|
576 |
with open(transcript_path, "w") as f:
|
577 |
f.write(text)
|
578 |
summary_text = text[:4096] if len(text) > 4096 else text
|
579 |
-
|
580 |
-
if len(text) > 100:
|
581 |
-
summary = summarizer(summary_text, max_length=200, min_length=50, do_sample=False)[0]['summary_text']
|
582 |
-
else:
|
583 |
-
summary = "Transcript too short for meaningful summarization."
|
584 |
-
except Exception as e:
|
585 |
-
summary = "Summarization failed due to an error."
|
586 |
-
print(f"Summarization error: {e}")
|
587 |
entities = extract_named_entities(text)
|
588 |
risk_analysis = analyze_risk_enhanced(text)
|
589 |
clauses = analyze_contract_clauses(text)
|
@@ -616,7 +484,7 @@ async def get_transcript(transcript_id: str):
|
|
616 |
async def legal_chatbot_api(query: str = Form(...), task_id: str = Form(...)):
|
617 |
document_context = load_document_context(task_id)
|
618 |
if not document_context:
|
619 |
-
return {"response": "
|
620 |
response = legal_chatbot(query, document_context)
|
621 |
return {"response": response, "chat_history": chat_history[-5:]}
|
622 |
|
@@ -646,95 +514,129 @@ def setup_ngrok():
|
|
646 |
try:
|
647 |
tunnels = ngrok.get_tunnels()
|
648 |
if not tunnels:
|
649 |
-
print("
|
650 |
ngrok_tunnel = ngrok.connect(8500, "http")
|
651 |
print(f"✅ Reconnected. New URL: {ngrok_tunnel.public_url}")
|
652 |
except Exception as e:
|
653 |
-
print(f"
|
654 |
Thread(target=keep_alive, daemon=True).start()
|
655 |
return public_url
|
656 |
except Exception as e:
|
657 |
-
print(f"
|
658 |
return None
|
659 |
|
660 |
-
|
661 |
-
|
|
|
|
|
|
|
|
|
662 |
try:
|
663 |
text = load_document_context(task_id)
|
664 |
if not text:
|
665 |
raise HTTPException(status_code=404, detail="Document context not found")
|
666 |
-
|
667 |
-
|
668 |
-
|
669 |
-
|
670 |
-
|
671 |
-
|
672 |
-
plt.
|
673 |
-
plt.xlabel("Clause Type")
|
674 |
-
plt.ylabel("Confidence Score")
|
675 |
-
plt.title("Extracted Legal Clause Confidence Scores")
|
676 |
-
plt.xticks(rotation=45, ha="right")
|
677 |
-
plt.tight_layout()
|
678 |
-
bar_chart_path = os.path.join("static", f"clause_bar_chart_{task_id}.png")
|
679 |
-
plt.savefig(bar_chart_path)
|
680 |
plt.close()
|
681 |
-
return FileResponse(
|
682 |
except Exception as e:
|
683 |
-
raise HTTPException(status_code=500, detail=f"Error generating
|
684 |
|
685 |
-
@app.get("/
|
686 |
-
async def
|
687 |
try:
|
688 |
text = load_document_context(task_id)
|
689 |
if not text:
|
690 |
raise HTTPException(status_code=404, detail="Document context not found")
|
691 |
-
|
692 |
-
|
693 |
-
|
694 |
-
|
695 |
-
|
696 |
-
labels = list(clause_counter.keys())
|
697 |
-
sizes = list(clause_counter.values())
|
698 |
plt.figure(figsize=(6, 6))
|
699 |
-
|
700 |
-
|
701 |
-
|
702 |
-
|
703 |
-
plt.title("Clause Type Distribution")
|
704 |
-
plt.tight_layout()
|
705 |
-
donut_chart_path = os.path.join("static", f"clause_donut_chart_{task_id}.png")
|
706 |
-
plt.savefig(donut_chart_path)
|
707 |
plt.close()
|
708 |
-
return FileResponse(
|
709 |
except Exception as e:
|
710 |
-
raise HTTPException(status_code=500, detail=f"Error generating
|
711 |
|
712 |
-
@app.get("/
|
713 |
-
async def
|
714 |
try:
|
715 |
text = load_document_context(task_id)
|
716 |
if not text:
|
717 |
raise HTTPException(status_code=404, detail="Document context not found")
|
718 |
-
|
719 |
-
|
720 |
-
|
721 |
-
|
722 |
-
values = [c["confidence"] for c in clauses]
|
723 |
-
labels += labels[:1]
|
724 |
values += values[:1]
|
725 |
-
angles = np.linspace(0, 2 * np.pi, len(
|
726 |
angles += angles[:1]
|
727 |
fig, ax = plt.subplots(figsize=(6, 6), subplot_kw=dict(polar=True))
|
728 |
ax.plot(angles, values, 'o-', linewidth=2)
|
729 |
ax.fill(angles, values, alpha=0.25)
|
730 |
-
ax.set_thetagrids(np.degrees(angles[:-1]),
|
731 |
-
ax.set_title("Legal
|
732 |
-
radar_chart_path = os.path.join("static", f"
|
733 |
plt.savefig(radar_chart_path)
|
734 |
plt.close()
|
735 |
-
return FileResponse(radar_chart_path, media_type="image/png", filename=f"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
736 |
except Exception as e:
|
737 |
-
raise HTTPException(status_code=500, detail=f"Error generating
|
738 |
|
739 |
def run():
|
740 |
print("Starting FastAPI server...")
|
@@ -745,5 +647,5 @@ if __name__ == "__main__":
|
|
745 |
if public_url:
|
746 |
print(f"\n✅ Your API is publicly available at: {public_url}/docs\n")
|
747 |
else:
|
748 |
-
print("\n
|
749 |
run()
|
|
|
1 |
import os
|
2 |
os.environ["TRANSFORMERS_NO_FAST"] = "1" # Force use of slow tokenizers
|
|
|
3 |
|
4 |
import io
|
5 |
import torch
|
|
|
13 |
import json
|
14 |
import tempfile
|
15 |
from fastapi import FastAPI, UploadFile, File, HTTPException, Form, BackgroundTasks
|
16 |
+
from fastapi.responses import FileResponse, JSONResponse, HTMLResponse # Added HTMLResponse
|
17 |
from fastapi.middleware.cors import CORSMiddleware
|
18 |
from transformers import pipeline, AutoModelForQuestionAnswering, AutoTokenizer
|
19 |
from sentence_transformers import SentenceTransformer
|
|
|
27 |
# For asynchronous blocking calls
|
28 |
from starlette.concurrency import run_in_threadpool
|
29 |
|
30 |
+
# Import gensim for topic modeling
|
31 |
import gensim
|
32 |
from gensim import corpora, models
|
33 |
|
|
|
|
|
|
|
34 |
# Global cache for analysis results based on file hash
|
35 |
analysis_cache = {}
|
36 |
|
37 |
+
# Ensure compatibility with Google Colab
|
38 |
try:
|
39 |
from google.colab import drive
|
40 |
drive.mount('/content/drive')
|
|
|
45 |
os.makedirs("static", exist_ok=True)
|
46 |
os.makedirs("temp", exist_ok=True)
|
47 |
|
48 |
+
# Ensure GPU usage
|
49 |
device = "cuda" if torch.cuda.is_available() else "cpu"
|
50 |
|
51 |
# Initialize FastAPI
|
|
|
64 |
document_storage = {}
|
65 |
chat_history = []
|
66 |
|
67 |
+
# Function to store document context by task ID
|
68 |
def store_document_context(task_id, text):
|
69 |
document_storage[task_id] = text
|
70 |
return True
|
71 |
|
72 |
+
# Function to load document context by task ID
|
73 |
def load_document_context(task_id):
|
74 |
return document_storage.get(task_id, "")
|
75 |
|
76 |
+
# Utility to compute MD5 hash from file content
|
77 |
def compute_md5(content: bytes) -> str:
|
78 |
return hashlib.md5(content).hexdigest()
|
79 |
|
|
|
83 |
|
84 |
def fine_tune_cuad_model():
|
85 |
from datasets import load_dataset
|
86 |
+
import numpy as np
|
87 |
from transformers import Trainer, TrainingArguments, AutoModelForQuestionAnswering, AutoTokenizer
|
88 |
|
89 |
print("✅ Loading CUAD dataset for fine tuning...")
|
|
|
121 |
tokenized_examples["end_positions"] = []
|
122 |
for i, offsets in enumerate(offset_mapping):
|
123 |
input_ids = tokenized_examples["input_ids"][i]
|
124 |
+
cls_index = input_ids.index(tokenizer.cls_token_id)
|
|
|
|
|
|
|
125 |
sequence_ids = tokenized_examples.sequence_ids(i)
|
126 |
sample_index = sample_mapping[i]
|
127 |
answers = examples["answers"][sample_index]
|
|
|
132 |
start_char = answers["answer_start"][0]
|
133 |
end_char = start_char + len(answers["text"][0])
|
134 |
tokenized_start_index = 0
|
135 |
+
while sequence_ids[tokenized_start_index] != 1:
|
136 |
tokenized_start_index += 1
|
137 |
tokenized_end_index = len(input_ids) - 1
|
138 |
+
while sequence_ids[tokenized_end_index] != 1:
|
139 |
tokenized_end_index -= 1
|
140 |
+
if not (offsets[tokenized_start_index][0] <= start_char and offsets[tokenized_end_index][1] >= end_char):
|
|
|
|
|
|
|
141 |
tokenized_examples["start_positions"].append(cls_index)
|
142 |
tokenized_examples["end_positions"].append(cls_index)
|
143 |
else:
|
144 |
while tokenized_start_index < len(offsets) and offsets[tokenized_start_index][0] <= start_char:
|
145 |
tokenized_start_index += 1
|
146 |
+
tokenized_examples["start_positions"].append(tokenized_start_index - 1)
|
147 |
+
while offsets[tokenized_end_index][1] >= end_char:
|
|
|
148 |
tokenized_end_index -= 1
|
149 |
+
tokenized_examples["end_positions"].append(tokenized_end_index + 1)
|
|
|
150 |
return tokenized_examples
|
151 |
|
152 |
print("✅ Tokenizing dataset...")
|
|
|
190 |
#############################
|
191 |
|
192 |
try:
|
|
|
193 |
try:
|
194 |
nlp = spacy.load("en_core_web_sm")
|
195 |
except Exception:
|
196 |
spacy.cli.download("en_core_web_sm")
|
197 |
nlp = spacy.load("en_core_web_sm")
|
198 |
+
print("✅ Loading NLP models...")
|
199 |
+
from transformers import PegasusTokenizer
|
|
|
200 |
summarizer = pipeline(
|
201 |
"summarization",
|
202 |
+
model="nsi319/legal-pegasus",
|
203 |
+
tokenizer=PegasusTokenizer.from_pretrained("nsi319/legal-pegasus", use_fast=False),
|
204 |
+
device=0 if torch.cuda.is_available() else -1
|
|
|
|
|
|
|
|
|
|
|
205 |
)
|
206 |
+
# Optionally convert summarizer model to FP16 for faster inference on GPU
|
207 |
+
if device == "cuda":
|
208 |
+
summarizer.model.half()
|
209 |
|
|
|
210 |
embedding_model = SentenceTransformer("all-mpnet-base-v2", device=device)
|
211 |
+
ner_model = pipeline("ner", model="dslim/bert-base-NER", device=0 if torch.cuda.is_available() else -1)
|
|
|
|
|
|
|
212 |
speech_to_text = pipeline("automatic-speech-recognition", model="openai/whisper-medium", chunk_length_s=30,
|
213 |
+
device_map="auto" if torch.cuda.is_available() else "cpu")
|
|
|
|
|
214 |
if os.path.exists("fine_tuned_legal_qa"):
|
215 |
print("✅ Loading fine-tuned CUAD QA model from fine_tuned_legal_qa...")
|
216 |
cuad_tokenizer = AutoTokenizer.from_pretrained("fine_tuned_legal_qa")
|
217 |
from transformers import AutoModelForQuestionAnswering
|
218 |
cuad_model = AutoModelForQuestionAnswering.from_pretrained("fine_tuned_legal_qa")
|
219 |
cuad_model.to(device)
|
220 |
+
if device == "cuda":
|
221 |
+
cuad_model.half()
|
222 |
else:
|
223 |
+
print("⚠ Fine-tuned QA model not found. Starting fine tuning on CUAD QA dataset. This may take a while...")
|
224 |
cuad_tokenizer, cuad_model = fine_tune_cuad_model()
|
225 |
cuad_model.to(device)
|
226 |
+
print("✅ All models loaded successfully")
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
227 |
except Exception as e:
|
228 |
+
print(f"⚠ Error loading models: {str(e)}")
|
229 |
raise RuntimeError(f"Error loading models: {str(e)}")
|
230 |
|
231 |
+
from transformers import pipeline
|
232 |
+
qa_model = pipeline("question-answering", model="deepset/roberta-base-squad2")
|
233 |
+
|
234 |
+
sentiment_pipeline = pipeline("sentiment-analysis", model="distilbert-base-uncased-finetuned-sst-2-english", device=0 if torch.cuda.is_available() else -1)
|
235 |
|
236 |
def legal_chatbot(user_input, context):
|
237 |
global chat_history
|
238 |
chat_history.append({"role": "user", "content": user_input})
|
239 |
+
response = qa_model(question=user_input, context=context)["answer"]
|
|
|
|
|
|
|
240 |
chat_history.append({"role": "assistant", "content": response})
|
241 |
return response
|
242 |
|
|
|
289 |
entities.extend([{"entity": ent.text, "label": ent.label_} for ent in doc.ents])
|
290 |
return entities
|
291 |
|
292 |
+
# -----------------------------
|
293 |
+
# Enhanced Risk Analysis Functions
|
294 |
+
# -----------------------------
|
295 |
|
296 |
def analyze_sentiment(text):
|
297 |
sentences = [sent.text for sent in nlp(text).sents]
|
|
|
318 |
enhanced["topics"] = analyze_topics(text, num_topics=5)
|
319 |
return enhanced
|
320 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
321 |
def analyze_risk_enhanced(text):
|
322 |
enhanced = get_enhanced_context_info(text)
|
323 |
avg_sentiment = enhanced["average_sentiment"]
|
324 |
risk_score = abs(avg_sentiment) if avg_sentiment < 0 else 0
|
325 |
+
return {"risk_score": risk_score, "average_sentiment": avg_sentiment, "topics": enhanced["topics"]}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
326 |
|
327 |
+
# -----------------------------
|
328 |
+
# Clause Detection (Chunk-Based)
|
329 |
+
# -----------------------------
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
330 |
|
331 |
def analyze_contract_clauses(text):
|
332 |
+
max_length = 512
|
333 |
+
step = 256
|
334 |
+
clauses_detected = []
|
335 |
try:
|
336 |
clause_types = list(cuad_model.config.id2label.values())
|
337 |
except Exception:
|
|
|
341 |
"Assignment", "Warranty", "Limitation of Liability", "Arbitration",
|
342 |
"IP Rights", "Force Majeure", "Revenue/Profit Sharing", "Audit Rights"
|
343 |
]
|
344 |
+
# Create chunks of the text
|
345 |
+
chunks = [text[i:i+max_length] for i in range(0, len(text), step) if i+step < len(text)]
|
346 |
+
for chunk in chunks:
|
347 |
+
inputs = cuad_tokenizer(chunk, return_tensors="pt", truncation=True, max_length=512).to(device)
|
348 |
+
with torch.no_grad():
|
349 |
+
outputs = cuad_model(**inputs)
|
350 |
+
predictions = torch.sigmoid(outputs.start_logits).cpu().numpy()[0]
|
351 |
+
for idx, confidence in enumerate(predictions):
|
352 |
+
if confidence > 0.5 and idx < len(clause_types):
|
353 |
+
clauses_detected.append({"type": clause_types[idx], "confidence": float(confidence)})
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
354 |
aggregated_clauses = {}
|
355 |
for clause in clauses_detected:
|
356 |
+
clause_type = clause["type"]
|
357 |
+
if clause_type not in aggregated_clauses or clause["confidence"] > aggregated_clauses[clause_type]["confidence"]:
|
358 |
+
aggregated_clauses[clause_type] = clause
|
359 |
return list(aggregated_clauses.values())
|
360 |
|
361 |
+
# -----------------------------
|
362 |
+
# Endpoints
|
363 |
+
# -----------------------------
|
364 |
|
365 |
@app.post("/analyze_legal_document")
|
366 |
async def analyze_legal_document(file: UploadFile = File(...)):
|
|
|
373 |
if not text:
|
374 |
return {"status": "error", "message": "No valid text found in the document."}
|
375 |
summary_text = text[:4096] if len(text) > 4096 else text
|
376 |
+
summary = summarizer(summary_text, max_length=200, min_length=50, do_sample=False)[0]['summary_text'] if len(text) > 100 else "Document too short for meaningful summarization."
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
377 |
entities = extract_named_entities(text)
|
378 |
risk_analysis = analyze_risk_enhanced(text)
|
379 |
clauses = analyze_contract_clauses(text)
|
|
|
411 |
with open(transcript_path, "w") as f:
|
412 |
f.write(text)
|
413 |
summary_text = text[:4096] if len(text) > 4096 else text
|
414 |
+
summary = summarizer(summary_text, max_length=200, min_length=50, do_sample=False)[0]['summary_text'] if len(text) > 100 else "Transcript too short for meaningful summarization."
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
415 |
entities = extract_named_entities(text)
|
416 |
risk_analysis = analyze_risk_enhanced(text)
|
417 |
clauses = analyze_contract_clauses(text)
|
|
|
451 |
with open(transcript_path, "w") as f:
|
452 |
f.write(text)
|
453 |
summary_text = text[:4096] if len(text) > 4096 else text
|
454 |
+
summary = summarizer(summary_text, max_length=200, min_length=50, do_sample=False)[0]['summary_text'] if len(text) > 100 else "Transcript too short for meaningful summarization."
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
455 |
entities = extract_named_entities(text)
|
456 |
risk_analysis = analyze_risk_enhanced(text)
|
457 |
clauses = analyze_contract_clauses(text)
|
|
|
484 |
async def legal_chatbot_api(query: str = Form(...), task_id: str = Form(...)):
|
485 |
document_context = load_document_context(task_id)
|
486 |
if not document_context:
|
487 |
+
return {"response": "⚠ No relevant document found for this task ID."}
|
488 |
response = legal_chatbot(query, document_context)
|
489 |
return {"response": response, "chat_history": chat_history[-5:]}
|
490 |
|
|
|
514 |
try:
|
515 |
tunnels = ngrok.get_tunnels()
|
516 |
if not tunnels:
|
517 |
+
print("⚠ Ngrok tunnel closed. Reconnecting...")
|
518 |
ngrok_tunnel = ngrok.connect(8500, "http")
|
519 |
print(f"✅ Reconnected. New URL: {ngrok_tunnel.public_url}")
|
520 |
except Exception as e:
|
521 |
+
print(f"⚠ Ngrok error: {e}")
|
522 |
Thread(target=keep_alive, daemon=True).start()
|
523 |
return public_url
|
524 |
except Exception as e:
|
525 |
+
print(f"⚠ Ngrok setup error: {e}")
|
526 |
return None
|
527 |
|
528 |
+
# ------------------------------
|
529 |
+
# Dynamic Visualization Endpoints
|
530 |
+
# ------------------------------
|
531 |
+
|
532 |
+
@app.get("/download_risk_chart")
|
533 |
+
async def download_risk_chart(task_id: str):
|
534 |
try:
|
535 |
text = load_document_context(task_id)
|
536 |
if not text:
|
537 |
raise HTTPException(status_code=404, detail="Document context not found")
|
538 |
+
risk_analysis = analyze_risk_enhanced(text)
|
539 |
+
plt.figure(figsize=(8, 5))
|
540 |
+
plt.bar(["Risk Score"], [risk_analysis["risk_score"]], color='red')
|
541 |
+
plt.ylabel("Risk Score")
|
542 |
+
plt.title("Legal Risk Assessment (Enhanced)")
|
543 |
+
risk_chart_path = os.path.join("static", f"risk_chart_{task_id}.png")
|
544 |
+
plt.savefig(risk_chart_path)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
545 |
plt.close()
|
546 |
+
return FileResponse(risk_chart_path, media_type="image/png", filename=f"risk_chart_{task_id}.png")
|
547 |
except Exception as e:
|
548 |
+
raise HTTPException(status_code=500, detail=f"Error generating risk chart: {str(e)}")
|
549 |
|
550 |
+
@app.get("/download_risk_pie_chart")
|
551 |
+
async def download_risk_pie_chart(task_id: str):
|
552 |
try:
|
553 |
text = load_document_context(task_id)
|
554 |
if not text:
|
555 |
raise HTTPException(status_code=404, detail="Document context not found")
|
556 |
+
risk_analysis = analyze_risk_enhanced(text)
|
557 |
+
labels = ["Risk", "No Risk"]
|
558 |
+
risk_value = risk_analysis["risk_score"]
|
559 |
+
risk_value = min(max(risk_value, 0), 1)
|
560 |
+
values = [risk_value, 1 - risk_value]
|
|
|
|
|
561 |
plt.figure(figsize=(6, 6))
|
562 |
+
plt.pie(values, labels=labels, autopct='%1.1f%%', startangle=90)
|
563 |
+
plt.title("Legal Risk Distribution (Enhanced)")
|
564 |
+
pie_chart_path = os.path.join("static", f"risk_pie_chart_{task_id}.png")
|
565 |
+
plt.savefig(pie_chart_path)
|
|
|
|
|
|
|
|
|
566 |
plt.close()
|
567 |
+
return FileResponse(pie_chart_path, media_type="image/png", filename=f"risk_pie_chart_{task_id}.png")
|
568 |
except Exception as e:
|
569 |
+
raise HTTPException(status_code=500, detail=f"Error generating pie chart: {str(e)}")
|
570 |
|
571 |
+
@app.get("/download_risk_radar_chart")
|
572 |
+
async def download_risk_radar_chart(task_id: str):
|
573 |
try:
|
574 |
text = load_document_context(task_id)
|
575 |
if not text:
|
576 |
raise HTTPException(status_code=404, detail="Document context not found")
|
577 |
+
risk_analysis = analyze_risk_enhanced(text)
|
578 |
+
categories = ["Average Sentiment", "Risk Score"]
|
579 |
+
values = [risk_analysis["average_sentiment"], risk_analysis["risk_score"]]
|
580 |
+
categories += categories[:1]
|
|
|
|
|
581 |
values += values[:1]
|
582 |
+
angles = np.linspace(0, 2 * np.pi, len(categories), endpoint=False).tolist()
|
583 |
angles += angles[:1]
|
584 |
fig, ax = plt.subplots(figsize=(6, 6), subplot_kw=dict(polar=True))
|
585 |
ax.plot(angles, values, 'o-', linewidth=2)
|
586 |
ax.fill(angles, values, alpha=0.25)
|
587 |
+
ax.set_thetagrids(np.degrees(angles[:-1]), ["Sentiment", "Risk"])
|
588 |
+
ax.set_title("Legal Risk Radar Chart (Enhanced)", y=1.1)
|
589 |
+
radar_chart_path = os.path.join("static", f"risk_radar_chart_{task_id}.png")
|
590 |
plt.savefig(radar_chart_path)
|
591 |
plt.close()
|
592 |
+
return FileResponse(radar_chart_path, media_type="image/png", filename=f"risk_radar_chart_{task_id}.png")
|
593 |
+
except Exception as e:
|
594 |
+
raise HTTPException(status_code=500, detail=f"Error generating radar chart: {str(e)}")
|
595 |
+
|
596 |
+
@app.get("/download_risk_trend_chart")
|
597 |
+
async def download_risk_trend_chart(task_id: str):
|
598 |
+
try:
|
599 |
+
text = load_document_context(task_id)
|
600 |
+
if not text:
|
601 |
+
raise HTTPException(status_code=404, detail="Document context not found")
|
602 |
+
words = text.split()
|
603 |
+
segments = np.array_split(words, 4)
|
604 |
+
segment_texts = [" ".join(segment) for segment in segments]
|
605 |
+
trend_scores = []
|
606 |
+
for segment in segment_texts:
|
607 |
+
risk = analyze_risk_enhanced(segment)
|
608 |
+
trend_scores.append(risk["risk_score"])
|
609 |
+
segments_labels = [f"Segment {i+1}" for i in range(len(segment_texts))]
|
610 |
+
plt.figure(figsize=(10, 6))
|
611 |
+
plt.plot(segments_labels, trend_scores, marker='o')
|
612 |
+
plt.xlabel("Document Segments")
|
613 |
+
plt.ylabel("Risk Score")
|
614 |
+
plt.title("Dynamic Legal Risk Trends (Enhanced)")
|
615 |
+
plt.xticks(rotation=45)
|
616 |
+
trend_chart_path = os.path.join("static", f"risk_trend_chart_{task_id}.png")
|
617 |
+
plt.savefig(trend_chart_path, bbox_inches="tight")
|
618 |
+
plt.close()
|
619 |
+
return FileResponse(trend_chart_path, media_type="image/png", filename=f"risk_trend_chart_{task_id}.png")
|
620 |
+
except Exception as e:
|
621 |
+
raise HTTPException(status_code=500, detail=f"Error generating trend chart: {str(e)}")
|
622 |
+
|
623 |
+
@app.get("/interactive_risk_chart", response_class=HTMLResponse)
|
624 |
+
async def interactive_risk_chart(task_id: str):
|
625 |
+
try:
|
626 |
+
import pandas as pd
|
627 |
+
import plotly.express as px
|
628 |
+
text = load_document_context(task_id)
|
629 |
+
if not text:
|
630 |
+
raise HTTPException(status_code=404, detail="Document context not found")
|
631 |
+
risk_analysis = analyze_risk_enhanced(text)
|
632 |
+
df = pd.DataFrame({
|
633 |
+
"Metric": ["Average Sentiment", "Risk Score"],
|
634 |
+
"Value": [risk_analysis["average_sentiment"], risk_analysis["risk_score"]]
|
635 |
+
})
|
636 |
+
fig = px.bar(df, x="Metric", y="Value", title="Interactive Enhanced Legal Risk Assessment")
|
637 |
+
return fig.to_html()
|
638 |
except Exception as e:
|
639 |
+
raise HTTPException(status_code=500, detail=f"Error generating interactive chart: {str(e)}")
|
640 |
|
641 |
def run():
|
642 |
print("Starting FastAPI server...")
|
|
|
647 |
if public_url:
|
648 |
print(f"\n✅ Your API is publicly available at: {public_url}/docs\n")
|
649 |
else:
|
650 |
+
print("\n⚠ Ngrok setup failed. API will only be available locally.\n")
|
651 |
run()
|