Update app.py

app.py

@@ -1,39 +1,29 @@
-import os
-import getpass
-
-import sentence_transformers
-
 import streamlit as st
-
-
-# Streamlit UI
-st.title("🔍 IBM Watson RAG Chatbot")
-
-# User input in Streamlit
-question = st.text_input("Enter your question:")
-
-if question:
-    # Retrieve relevant grounding context
-    grounding = RAG_proximity_search(question)
-
-    # Format the question with retrieved context
-    formatted_question = f"""<|start_of_role|>user<|end_of_role|>Use the following pieces of context to answer the question.
-
-{grounding}
-
-Question: {question}<|end_of_text|>
-<|start_of_role|>assistant<|end_of_role|>"""
-
-    # Placeholder for a prompt input (Optional)
-    prompt_input = ""  # Set this dynamically if needed
-    prompt = f"""{prompt_input}{formatted_question}"""
-
-    # Simulated AI response (Replace with actual model call)
-    generated_response = f"AI Response based on: {prompt}"
-
-    # Display results
-    st.subheader("📌 Retrieved Context")
-    st.write(grounding)
-
-    st.subheader("🤖 AI Response")
-    st.write(generated_response)
+from RAG import RAGinit, RAG_proximity_search
+
+# Initialize everything once at startup
+client, model, emb, chroma_collection, vector_index_properties, top_n = RAGinit()
+
+def main():
+    st.title("RAG-based QA App")
+
+    question = st.text_input("Ask a question:")
+    
+    if st.button("Search"):
+        if question.strip():
+            answer = RAG_proximity_search(
+                question,
+                client,
+                model,
+                emb,
+                chroma_collection,
+                vector_index_properties,
+                top_n
+            )
+            st.markdown("**Answer:**")
+            st.write(answer)
+        else:
+            st.warning("Please enter a question before searching.")
+
+if __name__ == "__main__":
+    main()