Spaces:
Running
Running
File size: 16,224 Bytes
9d8df86 38b3cc5 9d8df86 ff1b4d3 897fb70 38b3cc5 9d8df86 897fb70 ff1b4d3 897fb70 ff1b4d3 897fb70 ff1b4d3 897fb70 9d8df86 38b3cc5 9d8df86 38b3cc5 9d8df86 897fb70 9d8df86 897fb70 9d8df86 897fb70 9d8df86 ff1b4d3 9d8df86 ff1b4d3 9d8df86 ff1b4d3 9d8df86 897fb70 9d8df86 897fb70 38b3cc5 897fb70 9d8df86 897fb70 9d8df86 38b3cc5 897fb70 9d8df86 897fb70 38b3cc5 9d8df86 ff1b4d3 897fb70 ff1b4d3 897fb70 ff1b4d3 9d8df86 ff1b4d3 9d8df86 ff1b4d3 9d8df86 ff1b4d3 9d8df86 897fb70 9d8df86 897fb70 ff1b4d3 897fb70 ff1b4d3 9d8df86 897fb70 9d8df86 ff1b4d3 9d8df86 ff1b4d3 9d8df86 897fb70 9d8df86 897fb70 bfc0f42 897fb70 9d8df86 897fb70 9d8df86 897fb70 9d8df86 897fb70 bfc0f42 897fb70 9d8df86 897fb70 9d8df86 ff1b4d3 9d8df86 897fb70 ff1b4d3 897fb70 ff1b4d3 897fb70 ff1b4d3 9d8df86 ff1b4d3 897fb70 ff1b4d3 9d8df86 ff1b4d3 897fb70 ff1b4d3 9d8df86 38b3cc5 9d8df86 38b3cc5 9d8df86 |
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 |
import os
import re
import tempfile
import requests
import gradio as gr
from PyPDF2 import PdfReader
import logging
import webbrowser
from huggingface_hub import InferenceClient
from typing import Dict, List, Optional, Tuple
import time
# Set up logging
logging.basicConfig(level=logging.INFO, format='%(asctime)s - %(levelname)s - %(message)s')
# Constants
CONTEXT_SIZES = {
"4K": 4000,
"8K": 8000,
"32K": 32000,
"128K": 128000,
"200K": 200000
}
class ModelRegistry:
def __init__(self):
self.hf_models = {
"Phi-3 Mini 128k": "microsoft/Phi-3-mini-128k-instruct",
"Custom Model": ""
}
self.groq_models = self._fetch_groq_models()
def _fetch_groq_models(self) -> Dict[str, str]:
"""Fetch available Groq models"""
try:
headers = {
"Authorization": f"Bearer {os.getenv('GROQ_API_KEY')}",
"Content-Type": "application/json"
}
response = requests.get("https://api.groq.com/openai/v1/models", headers=headers)
if response.status_code == 200:
models = response.json().get("data", [])
return {model["id"]: model["id"] for model in models}
else:
logging.error(f"Failed to fetch Groq models: {response.status_code}")
return self._get_default_groq_models()
except Exception as e:
logging.error(f"Error fetching Groq models: {e}")
return self._get_default_groq_models()
def _get_default_groq_models(self) -> Dict[str, str]:
"""Return default Groq models when API is unavailable"""
return {
"llama-3.1-70b-versatile": "llama-3.1-70b-versatile",
"mixtral-8x7b-32768": "mixtral-8x7b-32768",
"llama-3.1-8b-instant": "llama-3.1-8b-instant"
}
def refresh_groq_models(self) -> Dict[str, str]:
"""Refresh the list of available Groq models"""
self.groq_models = self._fetch_groq_models()
return self.groq_models
# Initialize model registry
model_registry = ModelRegistry()
def extract_text_from_pdf(pdf_path: str) -> str:
"""Extract text content from PDF file."""
try:
reader = PdfReader(pdf_path)
text = ""
for page_num, page in enumerate(reader.pages, start=1):
page_text = page.extract_text()
if page_text:
text += page_text + "\n"
else:
logging.warning(f"No text found on page {page_num}.")
if not text.strip():
return "Error: No extractable text found in the PDF."
return text
except Exception as e:
logging.error(f"Error reading PDF file: {e}")
return f"Error reading PDF file: {e}"
def format_content(text: str, format_type: str) -> str:
"""Format extracted text according to specified format."""
if format_type == 'txt':
return text
elif format_type == 'md':
paragraphs = text.split('\n\n')
return '\n\n'.join(paragraphs)
elif format_type == 'html':
paragraphs = text.split('\n\n')
return ''.join([f'<p>{para.strip()}</p>' for para in paragraphs if para.strip()])
else:
logging.error(f"Unsupported format: {format_type}")
return f"Unsupported format: {format_type}"
def split_into_snippets(text: str, context_size: int) -> List[str]:
"""Split text into manageable snippets based on context size."""
sentences = re.split(r'(?<=[.!?]) +', text)
snippets = []
current_snippet = ""
for sentence in sentences:
if len(current_snippet) + len(sentence) + 1 > context_size:
if current_snippet:
snippets.append(current_snippet.strip())
current_snippet = sentence + " "
else:
snippets.append(sentence.strip())
current_snippet = ""
else:
current_snippet += sentence + " "
if current_snippet.strip():
snippets.append(current_snippet.strip())
return snippets
def build_prompts(snippets: List[str], prompt_instruction: str, custom_prompt: Optional[str], snippet_num: Optional[int] = None) -> str:
"""Build formatted prompts from text snippets."""
if snippet_num is not None:
if 1 <= snippet_num <= len(snippets):
selected_snippets = [snippets[snippet_num - 1]]
else:
return f"Error: Invalid snippet number. Please choose between 1 and {len(snippets)}."
else:
selected_snippets = snippets
prompts = []
base_prompt = custom_prompt if custom_prompt else prompt_instruction
for idx, snippet in enumerate(selected_snippets, start=1):
if len(selected_snippets) > 1:
prompt_header = f"{base_prompt} Part {idx} of {len(selected_snippets)}: ---\n"
else:
prompt_header = f"{base_prompt} ---\n"
framed_prompt = f"{prompt_header}{snippet}\n---"
prompts.append(framed_prompt)
return "\n\n".join(prompts)
def send_to_hf_inference(prompt: str, model_name: str, api_key: str) -> str:
"""Send prompt to HuggingFace using Inference API"""
try:
client = InferenceClient(api_key=api_key)
messages = [{"role": "user", "content": prompt}]
completion = client.chat.completions.create(
model=model_name,
messages=messages,
max_tokens=500
)
return completion.choices[0].message.content
except Exception as e:
logging.error(f"Error with HF inference: {e}")
return f"Error with HF inference: {e}"
def send_to_groq(prompt: str, model_name: str, api_key: str) -> str:
"""Send prompt to Groq API"""
try:
headers = {
"Authorization": f"Bearer {api_key}",
"Content-Type": "application/json"
}
data = {
"model": model_name,
"messages": [{"role": "user", "content": prompt}]
}
response = requests.post(
"https://api.groq.com/openai/v1/chat/completions",
headers=headers,
json=data
)
return response.json()["choices"][0]["message"]["content"]
except Exception as e:
logging.error(f"Error with Groq API: {e}")
return f"Error with Groq API: {e}"
def copy_to_clipboard(text: str) -> str:
"""Copy text to clipboard"""
return "Text copied to clipboard!"
def open_chatgpt() -> str:
"""Open ChatGPT in browser"""
webbrowser.open('https://chat.openai.com/')
return "Opening ChatGPT in browser..."
def process_pdf(pdf, fmt, ctx_size, snippet_num, prompt, model_selection,
hf_model_choice, hf_custom_model, hf_api_key,
groq_model_choice, groq_api_key) -> Tuple[str, str, str, List[str]]:
"""Process PDF and generate summary"""
try:
if not pdf:
return "Please upload a PDF file.", "", "", []
# Extract text
text = extract_text_from_pdf(pdf.name)
if text.startswith("Error"):
return text, "", "", []
# Format content
formatted_text = format_content(text, fmt)
# Split into snippets
snippets = split_into_snippets(formatted_text, ctx_size)
# Build prompts
default_prompt = "Summarize the following text:"
full_prompt = build_prompts(snippets, default_prompt, prompt, snippet_num)
if isinstance(full_prompt, str) and full_prompt.startswith("Error"):
return full_prompt, "", "", []
# Process with selected model
if model_selection == "HuggingFace Inference":
if not hf_api_key:
return "HuggingFace API key required.", full_prompt, "", []
model_id = hf_custom_model if hf_model_choice == "Custom Model" else model_registry.hf_models[hf_model_choice]
summary = send_to_hf_inference(full_prompt, model_id, hf_api_key)
elif model_selection == "Groq API":
if not groq_api_key:
return "Groq API key required.", full_prompt, "", []
summary = send_to_groq(full_prompt, groq_model_choice, groq_api_key)
else: # OpenAI ChatGPT
summary = "Please use the Copy Prompt button and paste into ChatGPT."
# Save files for download
files_to_download = []
with tempfile.NamedTemporaryFile(delete=False, mode='w', suffix='.txt') as prompt_file:
prompt_file.write(full_prompt)
files_to_download.append(prompt_file.name)
if summary != "Please use the Copy Prompt button and paste into ChatGPT.":
with tempfile.NamedTemporaryFile(delete=False, mode='w', suffix='.txt') as summary_file:
summary_file.write(summary)
files_to_download.append(summary_file.name)
return "Processing complete!", full_prompt, summary, files_to_download
except Exception as e:
logging.error(f"Error processing PDF: {e}")
return f"Error processing PDF: {str(e)}", "", "", []
# Main Interface
with gr.Blocks(theme=gr.themes.Default()) as demo:
# Store context size value
context_size_value = gr.State(value=32000)
# Header
gr.Markdown("# π Smart PDF Summarizer")
gr.Markdown("Upload a PDF document and get AI-powered summaries using various AI models.")
# Main Content
with gr.Row():
# Left Column - Input Options
with gr.Column(scale=1):
pdf_input = gr.File(
label="π Upload PDF",
file_types=[".pdf"]
)
with gr.Row():
format_type = gr.Radio(
choices=["txt", "md", "html"],
value="txt",
label="π Output Format"
)
gr.Markdown("### Context Window Size")
with gr.Row():
context_buttons = []
for size_name, size_value in CONTEXT_SIZES.items():
btn = gr.Button(size_name)
context_buttons.append((btn, size_value))
context_size = gr.Slider(
minimum=1000,
maximum=200000,
step=1000,
value=32000,
label="π Custom Context Size"
)
snippet_number = gr.Number(
label="π’ Snippet Number",
value=1,
precision=0
)
custom_prompt = gr.Textbox(
label="βοΈ Custom Prompt",
placeholder="Enter your custom prompt here...",
lines=2
)
model_choice = gr.Radio(
choices=["OpenAI ChatGPT", "HuggingFace Inference", "Groq API"],
value="OpenAI ChatGPT",
label="π€ Model Selection"
)
with gr.Column(visible=False) as hf_options:
hf_model = gr.Dropdown(
choices=list(model_registry.hf_models.keys()),
label="π§ HuggingFace Model",
value="Phi-3 Mini 128k"
)
hf_custom_model = gr.Textbox(
label="Custom Model ID",
placeholder="Enter custom model ID...",
visible=False
)
hf_api_key = gr.Textbox(
label="π HuggingFace API Key",
type="password"
)
with gr.Column(visible=False) as groq_options:
groq_model = gr.Dropdown(
choices=list(model_registry.groq_models.keys()),
label="π§ Groq Model",
value=list(model_registry.groq_models.keys())[0]
)
groq_refresh_btn = gr.Button("π Refresh Models")
groq_api_key = gr.Textbox(
label="π Groq API Key",
type="password"
)
# Right Column - Output
with gr.Column(scale=1):
process_button = gr.Button("π Process PDF", variant="primary")
progress_status = gr.Textbox(
label="π Progress",
interactive=False
)
generated_prompt = gr.Textbox(
label="π Generated Prompt",
lines=10
)
with gr.Row():
copy_prompt_button = gr.Button("π Copy Prompt")
open_chatgpt_button = gr.Button("π Open ChatGPT")
summary_output = gr.Textbox(
label="π Summary",
lines=15
)
with gr.Row():
copy_summary_button = gr.Button("π Copy Summary")
download_files = gr.Files(
label="π₯ Download Files"
)
# Event Handlers
def update_context_size(size):
return gr.update(value=size)
def toggle_model_options(choice):
return (
gr.update(visible=choice == "HuggingFace Inference"),
gr.update(visible=choice == "Groq API")
)
def refresh_groq_models_list():
updated_models = model_registry.refresh_groq_models()
return gr.update(choices=list(updated_models.keys()))
def toggle_custom_model(model_name):
return gr.update(visible=model_name == "Custom Model")
# Connect event handlers
model_choice.change(
toggle_model_options,
inputs=[model_choice],
outputs=[hf_options, groq_options]
)
for btn, size_value in context_buttons:
btn.click(
lambda v=size_value: v, # Simplified to directly return the value
None,
context_size
)
hf_model.change(
toggle_custom_model,
inputs=[hf_model],
outputs=[hf_custom_model]
)
groq_refresh_btn.click(
refresh_groq_models_list,
outputs=[groq_model]
)
process_button.click(
process_pdf,
inputs=[
pdf_input,
format_type,
context_size,
snippet_number,
custom_prompt,
model_choice,
hf_model,
hf_custom_model,
hf_api_key,
groq_model,
groq_api_key
],
outputs=[
progress_status,
generated_prompt,
summary_output,
download_files
]
)
copy_prompt_button.click(
copy_to_clipboard,
inputs=[generated_prompt],
outputs=[progress_status]
)
copy_summary_button.click(
copy_to_clipboard,
inputs=[summary_output],
outputs=[progress_status]
)
open_chatgpt_button.click(
open_chatgpt,
outputs=[progress_status]
)
# Instructions
gr.Markdown("""
### π Instructions:
1. Upload a PDF document
2. Choose output format and context window size
3. Select snippet number (default: 1) or enter custom prompt
4. Select your preferred model:
- OpenAI ChatGPT: Manual copy/paste workflow
- HuggingFace Inference: Direct API integration
- Groq API: High-performance inference
5. Click 'Process PDF' to generate summary
6. Use 'Copy Prompt' and 'Open ChatGPT' for manual processing
7. Download generated files as needed
### βοΈ Features:
- Support for multiple PDF formats
- Flexible text formatting options
- Predefined context window sizes (4K to 200K)
- Multiple model integrations
- Copy to clipboard functionality
- Direct ChatGPT integration
- Downloadable outputs
""")
# Launch the interface
if __name__ == "__main__":
demo.launch(share=False, debug=True) |