pdf_summarizer.py

# -*- coding: utf-8 -*-
"""pdf_summarizer.ipynb

Automatically generated by Colab.

Original file is located at
    https://colab.research.google.com/drive/1XblbxoRxB4XOHixjGij789FPD9KjKdhi
"""

import os
import PyPDF2
import gradio as gr
from langchain_groq.chat_models import ChatGroq

# Set Groq API key securely
GROQ_API_KEY = os.getenv("GROQ_API_KEY")  # Fetch from environment variables

# Ensure API key is available
if not GROQ_API_KEY:
    raise ValueError("GROQ_API_KEY is not set. Add it in Hugging Face Secrets.")

# Initialize LLM (Mistral-8x7B)
llm = ChatGroq(model_name="mixtral-8x7b-32768")

def extract_text_from_pdf(pdf_file):
    """Extract text from a PDF file."""
    text = ""
    reader = PyPDF2.PdfReader(pdf_file)
    for page in reader.pages:
        page_text = page.extract_text()
        if page_text:
            text += page_text + "\n"
    return text

def summarize_text(text):
    """Summarize the text"""
    prompt = f"Summarize the following document:\n\n{text[:10000]}"  # Limit input size
    response = llm.predict(prompt)
    return response

def process_pdf(file):
    """Extract text and summarize PDF using Mistral-8x7B."""
    if file is None:
        return "No file uploaded."

    # Read file bytes and process it using PyPDF2
    pdf_reader = PyPDF2.PdfReader(file)
    text = ""
    for page in pdf_reader.pages:
        page_text = page.extract_text()
        if page_text:
            text += page_text + "\n"

    # Limit text size for API efficiency
    text = text[:10000] if len(text) > 10000 else text

    # Summarize
    summary = summarize_text(text)
    return summary

# Create Gradio Interface
interface = gr.Interface(
    fn=process_pdf,
    inputs=gr.File(label="Upload a PDF"),
    outputs="text",
    title="📄 PDF Summarizer",
    description="Upload a PDF file and get a summary"
)

# Run the app
interface.launch()