Spaces:
Sleeping
Sleeping
File size: 22,345 Bytes
f1b6b8c 671b09b f1b6b8c |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 |
import os
import re
import tempfile
import gradio as gr
from transformers import pipeline
import pdfplumber
from gtts import gTTS
import nltk
import numpy as np
from sklearn.feature_extraction.text import TfidfVectorizer
from pydub import AudioSegment
import faiss
from sentence_transformers import SentenceTransformer
from groq import Groq
from diffusers import StableDiffusionPipeline
import torch
from PIL import Image
# ==========================================================
# π§ NLTK Setup
# ==========================================================
for pkg in ["punkt", "punkt_tab"]:
try:
nltk.data.find(f"tokenizers/{pkg}")
except LookupError:
nltk.download(pkg)
# ==========================================================
# π Environment Setup
# ==========================================================
GROQ_API_KEY = os.getenv("GROQ_API_KEY", "")
# ==========================================================
# βοΈ Model Setup
# ==========================================================
DEVICE = "cuda" if torch.cuda.is_available() else "cpu"
print(f"Using device: {DEVICE}")
# Initialize models
print("Loading models... please wait β³")
# Summarization model
SUMMARIZER_MODEL = "facebook/bart-large-cnn"
try:
summarizer = pipeline("summarization", model=SUMMARIZER_MODEL)
print("β
Summarizer loaded successfully.")
except Exception as e:
print("β Summarizer load error:", e)
summarizer = None
# Embedding model for RAG
try:
embedder = SentenceTransformer('all-MiniLM-L6-v2')
print("β
Embedding model loaded successfully.")
except Exception as e:
print("β Embedding model load error:", e)
embedder = None
# Stable Diffusion for diagram generation
try:
if torch.cuda.is_available():
sd_pipe = StableDiffusionPipeline.from_pretrained(
"runwayml/stable-diffusion-v1-5",
torch_dtype=torch.float16,
safety_checker=None,
requires_safety_checker=False
)
sd_pipe = sd_pipe.to("cuda")
else:
sd_pipe = StableDiffusionPipeline.from_pretrained("runwayml/stable-diffusion-v1-5")
sd_pipe = sd_pipe.to("cpu")
print("β
Stable Diffusion loaded successfully.")
except Exception as e:
print("β Stable Diffusion load error:", e)
sd_pipe = None
# Groq client
try:
groq_client = Groq(api_key=GROQ_API_KEY) if GROQ_API_KEY else None
if groq_client:
print("β
Groq client initialized successfully.")
else:
print("β οΈ Groq API key not found. Chat functionality will be limited.")
except Exception as e:
print("β Groq client initialization error:", e)
groq_client = None
# ==========================================================
# π§© Utility Functions
# ==========================================================
def clean_text(text: str) -> str:
"""Clean extracted PDF text."""
text = re.sub(r'\r\n?', '\n', text)
text = re.sub(r'\n{2,}', '\n\n', text)
text = re.sub(r'References[\s\S]*', '', text, flags=re.IGNORECASE)
text = re.sub(r'[^\x00-\x7F]+', ' ', text)
text = re.sub(r'\s+', ' ', text)
return text.strip()
def extract_text_from_pdf(path: str) -> str:
"""Extract text from all pages of a PDF."""
try:
text = ""
with pdfplumber.open(path) as pdf:
for page in pdf.pages:
page_text = page.extract_text()
if page_text:
text += page_text + "\n\n"
return text.strip() if text.strip() else "No text extracted from PDF."
except Exception as e:
return f"Error extracting text: {e}"
def sentence_tokenize(text: str):
"""Split text into sentences."""
return [s.strip() for s in nltk.tokenize.sent_tokenize(text) if len(s.strip()) > 10]
def chunk_text(text: str, max_chars=1500):
"""Split text into chunks for summarization."""
sents = sentence_tokenize(text)
chunks, cur = [], ""
for s in sents:
if len(cur) + len(s) < max_chars:
cur += (" " if cur else "") + s
else:
chunks.append(cur)
cur = s
if cur:
chunks.append(cur)
return chunks
def extract_keywords_tfidf(text: str, top_k=8):
"""Extract keywords using TF-IDF."""
try:
paras = [p.strip() for p in re.split(r'\n{2,}', text) if len(p.strip()) > 0]
vectorizer = TfidfVectorizer(stop_words='english', ngram_range=(1, 2))
X = vectorizer.fit_transform(paras)
features = vectorizer.get_feature_names_out()
scores = np.asarray(X.mean(axis=0)).ravel()
idx = np.argsort(scores)[::-1][:top_k]
return [features[i] for i in idx]
except Exception:
return []
# ==========================================================
# βοΈ Adaptive Summarization
# ==========================================================
def summarize_long_text(text: str) -> str:
"""Adaptive summarization based on PDF length."""
if summarizer is None:
return "Summarization model unavailable."
text = clean_text(text)
L = len(text)
# Dynamic summarization scaling
if L < 1500:
max_len, min_len, chunk_size = 180, 60, 1400
elif L < 5000:
max_len, min_len, chunk_size = 250, 100, 1600
elif L < 15000:
max_len, min_len, chunk_size = 350, 150, 1800
else:
max_len, min_len, chunk_size = 500, 200, 2000
if L <= chunk_size:
return summarizer(text, max_length=max_len, min_length=min_len, do_sample=False)[0]["summary_text"]
parts = chunk_text(text, max_chars=chunk_size)[:6]
summaries = []
for p in parts:
try:
summaries.append(summarizer(p, max_length=200, min_length=80, do_sample=False)[0]["summary_text"])
except Exception:
continue
combined = " ".join(summaries)
final = summarizer(combined, max_length=max_len, min_length=min_len, do_sample=False)[0]["summary_text"]
return final
# ==========================================================
# πΌοΈ Diagram Generation with Stable Diffusion
# ==========================================================
def generate_diagram(summary: str, keywords: str) -> Image.Image:
"""Generate a diagram based on summary and keywords."""
if sd_pipe is None:
return None
try:
# Create a prompt for diagram generation
prompt = f"educational diagram, infographic style, clean and professional, illustrating: {summary[:500]}. Keywords: {keywords}"
# Generate image
with torch.no_grad():
if torch.cuda.is_available():
image = sd_pipe(
prompt,
num_inference_steps=25,
guidance_scale=7.5,
width=512,
height=512
).images[0]
else:
image = sd_pipe(
prompt,
num_inference_steps=15,
guidance_scale=7.5,
width=512,
height=512
).images[0]
return image
except Exception as e:
print(f"Diagram generation error: {e}")
return None
# ==========================================================
# π¬ RAG Chatbot Functions
# ==========================================================
class PDFChatBot:
def __init__(self):
self.vector_store = None
self.chunks = []
self.current_pdf_text = ""
self.is_processed = False
def process_pdf_for_chat(self, pdf_text: str):
"""Process PDF text for RAG system."""
if not pdf_text or pdf_text.startswith("Error") or pdf_text.startswith("No text"):
return False
self.current_pdf_text = clean_text(pdf_text)
# Chunk the text
self.chunks = self._create_chunks(self.current_pdf_text, chunk_size=500, overlap=50)
# Create embeddings
if embedder is not None and self.chunks:
embeddings = embedder.encode(self.chunks)
# Create FAISS index
dimension = embeddings.shape[1]
self.vector_store = faiss.IndexFlatIP(dimension)
# Normalize embeddings for cosine similarity
faiss.normalize_L2(embeddings)
self.vector_store.add(embeddings)
self.is_processed = True
return True
return False
def _create_chunks(self, text: str, chunk_size: int = 500, overlap: int = 50):
"""Create overlapping chunks of text."""
sentences = sentence_tokenize(text)
chunks = []
current_chunk = ""
for sentence in sentences:
if len(current_chunk) + len(sentence) <= chunk_size:
current_chunk += " " + sentence
else:
if current_chunk:
chunks.append(current_chunk.strip())
current_chunk = sentence
if current_chunk:
chunks.append(current_chunk.strip())
return chunks
def get_relevant_chunks(self, query: str, top_k: int = 3):
"""Retrieve relevant chunks for a query."""
if self.vector_store is None or not self.chunks:
return []
try:
# Encode query
query_embedding = embedder.encode([query])
faiss.normalize_L2(query_embedding)
# Search
scores, indices = self.vector_store.search(query_embedding, top_k)
# Return relevant chunks
relevant_chunks = []
for i, score in zip(indices[0], scores[0]):
if i < len(self.chunks) and score > 0.3: # similarity threshold
relevant_chunks.append(self.chunks[i])
return relevant_chunks
except Exception as e:
print(f"Error in retrieval: {e}")
return []
def generate_answer(self, query: str, chat_history):
"""Generate answer using RAG with Groq."""
if groq_client is None:
return "Groq API not available. Please set your GROQ_API_KEY in the Hugging Face Spaces secrets."
if not self.is_processed:
return "Please upload and process a PDF first. Go to the 'PDF Summarizer' tab to upload your PDF."
# Get relevant context
relevant_chunks = self.get_relevant_chunks(query)
if not relevant_chunks:
return "No relevant information found in the PDF for your question."
context = "\n\n".join(relevant_chunks[:3]) # Use top 3 chunks
# Create prompt
prompt = f"""Based on the following context from a PDF document, please answer the user's question.
Context:
{context}
Question: {query}
Please provide a helpful and accurate answer based only on the given context. If the context doesn't contain enough information to fully answer the question, please say so."""
try:
# Try different available Groq models
available_models = [
"llama-3.3-70b-versatile",
"llama-3.1-8b-instant",
"llama-3.2-3b-preview",
"llama-3.2-1b-preview",
"mixtral-8x7b-32768"
]
for model in available_models:
try:
completion = groq_client.chat.completions.create(
model=model,
messages=[{"role": "user", "content": prompt}],
temperature=0.7,
max_tokens=1024,
top_p=1,
stream=False
)
answer = completion.choices[0].message.content
return answer
except Exception as model_error:
print(f"Model {model} failed: {model_error}")
continue
return "All available models failed. Please check your Groq API access."
except Exception as e:
return f"Error generating answer: {str(e)}"
# Initialize chatbot
chatbot = PDFChatBot()
# ==========================================================
# π Text-to-Speech
# ==========================================================
def text_to_speech(text):
"""Convert text to speech and ensure WAV output."""
if not text:
return None
try:
# Temporary paths
mp3_path = tempfile.NamedTemporaryFile(delete=False, suffix=".mp3").name
wav_path = tempfile.NamedTemporaryFile(delete=False, suffix=".wav").name
# Generate TTS (MP3)
gTTS(text=text[:900], lang="en").save(mp3_path)
# Convert to WAV for browser playback
AudioSegment.from_mp3(mp3_path).export(wav_path, format="wav")
# Clean up MP3 file
os.unlink(mp3_path)
return wav_path
except Exception as e:
print("TTS error:", e)
return None
# ==========================================================
# π PDF Processing - Main Function
# ==========================================================
def process_pdf(pdf_file):
"""Main handler to process PDF - this will be shared across all tabs."""
if not pdf_file:
return "Please upload a PDF.", "", None, "", None, "No PDF uploaded"
text = extract_text_from_pdf(pdf_file)
if text.startswith("Error") or text.startswith("No text"):
return text, "", None, "", None, "Failed to extract text"
text = clean_text(text)
summary = summarize_long_text(text)
keywords = ", ".join(extract_keywords_tfidf(text))
audio = text_to_speech(summary)
# Generate diagram
diagram = generate_diagram(summary, keywords)
# Also process for chatbot
chatbot.process_pdf_for_chat(text)
# Return status message for chat tab
status_message = "β
PDF processed successfully! You can now chat with this PDF in the 'Chat with PDF' tab."
return text, summary, audio, keywords, diagram, status_message
# ==========================================================
# π Gradio Interface with Shared PDF State
# ==========================================================
def create_interface():
with gr.Blocks(
title="AI PDF Summarizer Pro",
theme=gr.themes.Soft()
) as demo:
gr.Markdown("""
# AI PDF Summarizer Pro
*Upload once, use everywhere across all tabs*
""")
# --- Main Tab: PDF Summarizer ---
with gr.Tab("π PDF Summarizer"):
with gr.Row():
with gr.Column(scale=1):
gr.Markdown("### Upload Your PDF")
gr.Markdown("Upload a PDF here and it will be automatically available in all other tabs.")
pdf_input = gr.File(
label="Upload PDF Document",
file_types=[".pdf"],
type="filepath"
)
process_btn = gr.Button(
"Process PDF",
variant="primary",
size="lg"
)
with gr.Column(scale=2):
with gr.Accordion("Extracted Text", open=False):
extracted_text = gr.Textbox(
label="",
lines=8,
interactive=False,
show_copy_button=True
)
with gr.Row():
with gr.Column():
summary_box = gr.Textbox(
label="AI Summary",
lines=4,
interactive=False,
show_copy_button=True
)
with gr.Column():
keywords_box = gr.Textbox(
label="Top Keywords",
lines=2,
interactive=False
)
with gr.Row():
with gr.Column():
audio_box = gr.Audio(
label="Summary Audio",
type="filepath",
interactive=False
)
with gr.Column():
diagram_box = gr.Image(
label="AI Generated Diagram",
interactive=False,
height=200
)
# Status message
status_display = gr.HTML(
value="<div>No PDF processed yet. Upload a PDF and click 'Process PDF'.</div>"
)
# --- Tab: AI Diagram Generator ---
with gr.Tab("πΌοΈ AI Diagram"):
with gr.Row():
with gr.Column():
gr.Markdown("### Create Diagram")
gr.Markdown("Create diagrams using the summary from your uploaded PDF or enter custom text.")
diagram_summary_input = gr.Textbox(
label="Summary Text",
lines=3,
placeholder="Text from your PDF summary will appear here after processing..."
)
diagram_keywords_input = gr.Textbox(
label="Keywords (optional)",
placeholder="Keywords from your PDF will appear here..."
)
generate_diagram_btn = gr.Button(
"Generate Diagram",
variant="primary"
)
with gr.Column():
gr.Markdown("### Generated Diagram")
diagram_output = gr.Image(
label="",
interactive=False,
height=400,
show_download_button=True
)
# --- Tab: Chat with PDF ---
with gr.Tab("π¬ Chat with PDF"):
with gr.Row():
with gr.Column(scale=1):
gr.Markdown("### Chat with Your PDF")
gr.Markdown("""
**Ask questions about your uploaded PDF**
Simply go to the **PDF Summarizer** tab, upload and process your PDF, then come back here to start chatting!
""")
# Display current PDF status
chat_status_display = gr.HTML(
value="<div>Please upload and process a PDF in the 'PDF Summarizer' tab first.</div>"
)
with gr.Column(scale=2):
chatbot_interface = gr.ChatInterface(
fn=chatbot.generate_answer,
title="Chat with Your PDF",
description="Ask questions about the content of your uploaded PDF document",
examples=[
"What is the main topic of this document?",
"Can you summarize the key points?",
"What are the most important findings?",
"Explain the methodology used",
"What conclusions does the author reach?"
]
)
# --- Tab: About ---
with gr.Tab("βΉοΈ About"):
gr.Markdown("""
## About AI PDF Summarizer Pro
**One PDF Upload, Multiple AI Features**
Upload your PDF once in the **PDF Summarizer** tab and use it across all features:
- **π PDF Summarizer**: Extract text, generate summaries, get keywords
- **πΌοΈ AI Diagram**: Create visual diagrams from your content
- **π¬ Chat with PDF**: Ask questions and get instant answers
### How it works:
1. Upload your PDF in the **PDF Summarizer** tab
2. Click **Process PDF**
3. The same PDF is automatically available in all other tabs
4. No need to re-upload - seamless experience!
### Powered by:
- Hugging Face Transformers
- Stable Diffusion
- Groq API
- FAISS Vector Search
### Setup Instructions:
For full functionality, add your Groq API key in Hugging Face Spaces secrets:
- Go to your Space settings
- Add a secret named `GROQ_API_KEY` with your Groq API key
""")
# --- Event Handlers ---
# Main PDF processing - updates all tabs
process_btn.click(
process_pdf,
inputs=[pdf_input],
outputs=[extracted_text, summary_box, audio_box, keywords_box, diagram_box, status_display]
).then(
# Update the diagram tab inputs with the generated summary and keywords
lambda summary, keywords: (summary, keywords),
inputs=[summary_box, keywords_box],
outputs=[diagram_summary_input, diagram_keywords_input]
).then(
# Update chat status
lambda: "<div>β
PDF processed successfully! You can now chat with your document.</div>",
outputs=[chat_status_display]
)
# Standalone diagram generation
generate_diagram_btn.click(
generate_diagram,
inputs=[diagram_summary_input, diagram_keywords_input],
outputs=[diagram_output]
)
return demo
# ==========================================================
# π Launch Application
# ==========================================================
if __name__ == "__main__":
print("Starting AI PDF Summarizer Pro Version")
print("Key Feature: Upload PDF once, use across all tabs!")
print("Loading AI models...")
print("β
Summarization Model")
print("β
Embedding Model")
print("β
Diagram Generation")
print("β
Chat Model")
demo = create_interface()
demo.launch(share=False) |