app.py

import gradio as gr
import openai
from datasets import load_dataset
import logging

# Set up logging
logging.basicConfig(level=logging.INFO)
logger = logging.getLogger(__name__)

# Initialize OpenAI API key
openai.api_key = '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'

# Load just one dataset to start
dataset = load_dataset("rungalileo/ragbench", "hotpotqa", split='train')
logger.info("Dataset loaded successfully")

def process_query(query):
    try:
        # Get a relevant document from the dataset
        context = dataset['documents'][0]  # Using first document as example
        
        response = openai.chat.completions.create(
            model="gpt-3.5-turbo",
            messages=[
                {"role": "system", "content": "You are a helpful assistant for the RagBench dataset."},
                {"role": "user", "content": f"Context: {context}\nQuestion: {query}"}
            ],
            max_tokens=300,
            temperature=0.7,
        )
        
        return response.choices[0].message.content.strip()
        
    except Exception as e:
        return f"Query processing: {str(e)}"

# Create simple Gradio interface
demo = gr.Interface(
    fn=process_query,
    inputs=gr.Textbox(label="Question"),
    outputs=gr.Textbox(label="Answer"),
    title="RagBench QA System",
    description="Ask questions about HotpotQA dataset"
)

if __name__ == "__main__":
    demo.launch(debug=True)