app.py

import gradio as gr
from transformers import pipeline
from pptx import Presentation
import re

# Create a text classification pipeline
classifier = pipeline("text-classification", model="Ahmed235/roberta_classification", tokenizer="Ahmed235/roberta_classification")
summarizer = pipeline("summarization", model="Falconsai/text_summarization")

def extract_text_from_pptx(file_path):
    presentation = Presentation(file_path)
    text = []
    for slide_number, slide in enumerate(presentation.slides, start=1):
        for shape in slide.shapes:
            if hasattr(shape, "text"):
                text.append(shape.text)
    return "\n".join(text)

def limit_text_length(text, max_length=512):
    # Truncate or limit the text length
    return text[:max_length]

def predict_pptx_content(file_path):
    try:
        extracted_text = extract_text_from_pptx(file_path)
        cleaned_text = re.sub(r'\s+', ' ', extracted_text)
        
        # Limit text length before classification
        limited_text = limit_text_length(cleaned_text)

        # Perform inference using the pipeline
        result = classifier(limited_text)

        predicted_label = result[0]['label']
        predicted_probability = result[0]['score']
        summary = summarizer(extracted_text, max_length=1000, min_length=30, do_sample=False)[0]['summary_text']
        
        return {
            "predicted_label": predicted_label,
            "evaluation": predicted_probability,
            "summary": summary
        }

    except Exception as e:
        # Log the error details
        print(f"Error in predict_pptx_content: {e}")
        return {"error": str(e)}

# Define the Gradio interface
iface = gr.Interface(
    fn=predict_pptx_content,
    inputs=gr.File(type="file", label="Upload PowerPoint (.pptx) file"),
    outputs=gr.Textbox("output"),
    live=False,
    title="<h1 style='color: lightgreen; text-align: center;'>HackTalk Analyzer</h1>",
)

# Deploy the Gradio interface
iface.launch(share=True)