app.py

import streamlit as st
from transformers import AutoModelForQuestionAnswering, AutoTokenizer
import torch

# Load model and tokenizer
@st.cache_resource
def load_model():
    model_path = 'C:\Users\neeli\Downloads\bert-tensorflow2-uncased-tf2-qa-v1' 
    model = AutoModelForQuestionAnswering.from_pretrained(model_path)
    tokenizer = AutoTokenizer.from_pretrained(model_path)
    return model, tokenizer

model, tokenizer = load_model()

# Function to get answer from question and context
def get_answer(question, context):
    inputs = tokenizer.encode_plus(question, context, return_tensors='pt')
    with torch.no_grad():
        outputs = model(**inputs)
    
    answer_start = torch.argmax(outputs.start_logits)
    answer_end = torch.argmax(outputs.end_logits) + 1
    answer = tokenizer.convert_tokens_to_string(tokenizer.convert_ids_to_tokens(inputs['input_ids'][0][answer_start:answer_end]))
    return answer

# Streamlit UI
st.title("Question Answering Application")

question = st.text_input("Enter your question:")
context = st.text_area("Enter context text:", height=200)

if st.button("Get Answer"):
    if question and context:
        answer = get_answer(question, context)
        st.write(f"**Answer:** {answer}")
    else:
        st.warning("Please enter both a question and context.")