adding model params and QnA comp to UI

app.py

@@ -1,4 +1,6 @@
 import streamlit as st
+from utils.retriever import retrieve_paragraphs
+from utils.generator import generate
 
 col_title, col_about = st.columns([8, 2])
 with col_title:
@@ -6,15 +8,65 @@ with col_title:
         "<h1 style='text-align:center;'> Montreal AI Decisions (MVP)</h1>",
         unsafe_allow_html=True
     )
-with col_about:
-    with st.expander("ℹ️ About this prototype"):
-        st.markdown(
-            """
-            - **Data**: Public decisions under the Montreal Protocol.  
-            - **Search**: Semantic (with a 0.3 cosine threshold).  
-            - **RAG**: Answers generated with top-5 semantic search results as context.  
-            ⚠️ For testing only; do not input sensitive info.  
-            AI answers may be incorrect or misleading.
-            """,
-            unsafe_allow_html=True
-        )
+
+async def chat_response(query):
+    """Generate chat response based on method and inputs"""
+    
+    try:
+        retrieved_paragraphs = retrieve_paragraphs(query)
+        context_retrieved = ast.literal_eval(retrieved_paragraphs)
+        
+        # Build list of only content, no metadata
+        context_retrieved_formatted = "||".join(doc['answer'] for doc in context_retrieved)
+        context_retrieved_lst = [doc['answer'] for doc in context_retrieved]
+        
+        # # Prepare HTML for displaying source documents
+        # docs_html = []
+        # for i, d in enumerate(context_retrieved, 1):
+        #     docs_html.append(make_html_source(d, i))
+        # docs_html = "".join(docs_html)
+        
+        # Generate response
+        response = await generate(query=query, context=context_retrieved_lst)
+        
+        # Add disclaimer to the response
+        response_with_disclaimer = BEGINNING_TEXT + response
+        # Log the interaction
+        # try:
+        #     chat_logger.log(
+        #         query=query,
+        #         answer=response,
+        #         retrieved_content=context_retrieved_lst,
+        #         request=request
+        #     )
+        # except Exception as e:
+        #     print(f"Logging error: {str(e)}")
+        
+
+        # Stream response character by character
+        displayed_response = ""
+        for i, char in enumerate(response_with_disclaimer):
+            displayed_response += char
+            
+            yield displayed_response
+            # Only add delay every few characters to avoid being too slow
+            if i % 3 == 0:
+                await asyncio.sleep(0.02)
+                
+    except Exception as e:
+        error_message = f"Error processing request: {str(e)}"
+        yield error_message
+
+# 10.1. Question input
+query = st.text_input(
+    label="Enter your question:",
+    key="query",
+    on_change=reset_page
+)
+
+# Only run search & display if user has entered something
+if not query.strip():
+    st.info("Please enter a question to see results.")
+    st.stop()
+else:
+    st.write_stream(chat_response(query))

model_params.cfg

@@ -0,0 +1,7 @@
+[generator]
+PROVIDER = huggingface
+MODEL = meta-llama/Meta-Llama-3-8B-Instruct
+MAX_TOKENS = 768
+TEMPERATURE = 0.2
+INFERENCE_PROVIDER = novita
+ORGANIZATION = GIZ