Spaces:
Sleeping
Sleeping
| import PyPDF2 | |
| import gradio as gr | |
| from smolagents import CodeAgent, HfApiModel, tool | |
| from tools.final_answer import FinalAnswerTool | |
| # Define the PDF text extraction tool | |
| def extract_text_from_pdf(pdf_path: str) -> str: | |
| """Extracts text from a given PDF file. | |
| Args: | |
| pdf_path: The path to the PDF file. | |
| """ | |
| try: | |
| with open(pdf_path, "rb") as file: | |
| reader = PyPDF2.PdfReader(file) | |
| text = "\n".join([page.extract_text() for page in reader.pages if page.extract_text()]) | |
| return text if text else "No text found in the PDF." | |
| except Exception as e: | |
| return f"Error extracting text from PDF: {str(e)}" | |
| # Define the PDF summarization tool | |
| def summarize_text(text: str) -> str: | |
| """Summarizes the extracted text using the AI model. | |
| Args: | |
| text: The extracted text from the PDF. | |
| """ | |
| max_input_tokens = 15000 # ✅ Adjust input limit to stay within model constraints | |
| # Truncate text if it's too long | |
| truncated_text = text[:max_input_tokens] | |
| prompt = f"Summarize the following document:\n\n{truncated_text}" | |
| # ✅ Correct input format for HfApiModel | |
| response = model.__call__([{"role": "user", "content": prompt}]) | |
| return response | |
| # Initialize AI Model | |
| model = HfApiModel( | |
| max_tokens=1024, # Reduced token limit for summaries | |
| temperature=0.5, | |
| model_id='Qwen/Qwen2.5-Coder-32B-Instruct', | |
| custom_role_conversions=None, | |
| ) | |
| # Initialize Final Answer Tool | |
| final_answer = FinalAnswerTool() | |
| # Create AI Agent with PDF extraction and summarization tools | |
| agent = CodeAgent( | |
| model=model, | |
| tools=[final_answer, extract_text_from_pdf, summarize_text], # Added summarization tool | |
| max_steps=6, | |
| verbosity_level=1, | |
| grammar=None, | |
| planning_interval=None, | |
| name="PDF Summary Agent", | |
| description="An AI agent that extracts and summarizes text from PDFs.", | |
| prompt_templates=None | |
| ) | |
| # Define Gradio function to handle PDF upload and summarization | |
| def extract_and_summarize_pdf(file_path): | |
| extracted_text = extract_text_from_pdf(file_path) | |
| if "Error" in extracted_text: | |
| return extracted_text, "No summary available." | |
| summary = summarize_text(extracted_text) | |
| return extracted_text, summary | |
| # Gradio UI Setup | |
| with gr.Blocks() as ui: | |
| gr.Markdown("# 📄 PDF Summarizer Agent") | |
| gr.Markdown("### Upload a PDF file, and the agent will extract and summarize its content.") | |
| with gr.Row(): | |
| pdf_input = gr.File(type="filepath", label="Upload PDF") # ✅ Fixed type | |
| output_text = gr.Textbox(label="Extracted Text", interactive=False) | |
| summary_output = gr.Textbox(label="Summary", interactive=False) | |
| extract_button = gr.Button("Extract & Summarize") | |
| extract_button.click(extract_and_summarize_pdf, inputs=pdf_input, outputs=[output_text, summary_output]) | |
| # Launch Gradio UI | |
| ui.launch() | |