app.py

import gradio as gr
from transformers import pipeline
from PyPDF2 import PdfReader
import os

# Load a summarization model
summarizer = pipeline("summarization", model="facebook/bart-large-cnn")

# Function to summarize text
def summarize_text(text, max_length, min_length):
    summary = summarizer(text, max_length=int(max_length), min_length=int(min_length), do_sample=False)
    return summary[0]['summary_text']

# Function to summarize uploaded documents
def summarize_document(file, max_length, min_length):
    # Read the content of the file
    if file.name.endswith(".pdf"):
        reader = PdfReader(file.name)
        text = ""
        for page in reader.pages:
            text += page.extract_text()
    else:
        text = file.read().decode("utf-8")
    
    # Summarize the extracted text
    if len(text) > 1024:  # Summarizer has input size limitations
        text = text[:1024]
    
    summary = summarize_text(text, max_length, min_length)
    return summary

# Gradio Interface
with gr.Blocks() as interface:
    gr.Markdown("# Document Summarizer")
    
    with gr.Tab("Text Summarization"):
        input_text = gr.Textbox(lines=10, placeholder="Enter the text to summarize...", label="Input Text")
        max_len = gr.Slider(50, 300, value=130, label="Max Summary Length")
        min_len = gr.Slider(10, 50, value=30, label="Min Summary Length")
        output_text = gr.Textbox(lines=5, label="Summary")
        summarize_button = gr.Button("Summarize")
        
        summarize_button.click(summarize_text, [input_text, max_len, min_len], output_text)
    
    with gr.Tab("Document Summarization"):
        input_file = gr.File(label="Upload a Document (PDF or TXT)")
        max_len_doc = gr.Slider(50, 300, value=130, label="Max Summary Length")
        min_len_doc = gr.Slider(10, 50, value=30, label="Min Summary Length")
        output_doc = gr.Textbox(lines=5, label="Summary")
        summarize_doc_button = gr.Button("Summarize Document")
        
        summarize_doc_button.click(summarize_document, [input_file, max_len_doc, min_len_doc], output_doc)

# Launch the app
if __name__ == "__main__":
    interface.launch()