Update app.py

app.py

@@ -1,31 +1,34 @@
 import streamlit as st
 from transformers import pipeline
 
-# Load the conversational pipeline
-model_name = "./QAModel"
-chatbot_pipeline = pipeline("conversational", model=model_name, tokenizer=model_name)
+# Define the path to the saved model
+model_path = './QAModel'  # Path to your fine-tuned model
+
+# Load the question-answering pipeline
+qa_pipeline = pipeline("question-answering", model=model_path, tokenizer=model_path)
 
 # Set the title for the Streamlit app
-st.title("Movie Trivia Chatbot")
+st.title("Movie Trivia Question Answering")
 
-# Text input for the user
-user_input = st.text_area("Ask a movie trivia question:")
+# Text inputs for the user
+context = st.text_area("Enter the context (movie-related text):")
+question = st.text_area("Enter your question:")
 
-def get_response(user_input):
-    # Generate response
-    conversation = chatbot_pipeline(user_input)
-    return conversation[0]['generated_text']
+def generate_answer(question, context):
+    # Perform question answering
+    result = qa_pipeline(question=question, context=context)
+    return result['answer']
 
 if st.button("Get Answer"):
-    if user_input:
-        response = get_response(user_input)
-        # Display the response
+    if context and question:
+        generated_answer = generate_answer(question, context)
+        # Display the generated answer
         st.subheader("Answer")
-        st.write(response)
+        st.write(generated_answer)
     else:
-        st.warning("Please enter a question.")
+        st.warning("Please enter both context and question.")
 
 # Optionally, add instructions or information about the app
 st.write("""
-    Enter a movie-related question above. The chatbot will provide an answer based on its training.
+    Enter a movie-related context and a question related to the context above. The model will provide the answer based on the context provided.
     """)