Update app.py

app.py

@@ -150,19 +150,90 @@ if query:
     with st.spinner("🔄 Retrieving relevant context..."):
         retriever = st.session_state.vector_store.as_retriever(search_type="similarity", search_kwargs={"k": 5})
         retrieved_docs = retriever.invoke(query)
-        context = [d.page_content for d in retrieved_docs]
-        st.success("✅ Context retrieved successfully!")
+
+        # Debugging: Show retrieved documents
+        st.write("🔹 Retrieved Documents:", retrieved_docs)
+
+        if not retrieved_docs:
+            st.error("❌ No relevant documents retrieved! Try a different query.")
+        else:
+            # Ensure extracted content is formatted correctly
+            context = [d.page_content for d in retrieved_docs]
+
+            if isinstance(context, list):  # Convert list to string for LLMChain
+                context_str = "\n".join(context)
+            else:
+                context_str = str(context)
+
+            st.success("✅ Context retrieved successfully!")
+            st.write("🔹 Extracted Context:", context_str)
 
     # ----------------- Run Individual Chains Explicitly -----------------
-    context_relevancy_chain = LLMChain(llm=llm_judge, prompt=PromptTemplate(input_variables=["retriever_query", "context"], template=relevancy_prompt), output_key="relevancy_response")
-    relevant_context_chain = LLMChain(llm=llm_judge, prompt=PromptTemplate(input_variables=["relevancy_response"], template=relevant_context_picker_prompt), output_key="context_number")
-    relevant_contexts_chain = LLMChain(llm=llm_judge, prompt=PromptTemplate(input_variables=["context_number", "context"], template=response_synth), output_key="relevant_contexts")
-    response_chain = LLMChain(llm=rag_llm, prompt=PromptTemplate(input_variables=["query", "context"], template=rag_prompt), output_key="final_response")
 
-    response_crisis = context_relevancy_chain.invoke({"context": context, "retriever_query": query})
-    relevant_response = relevant_context_chain.invoke({"relevancy_response": response_crisis["relevancy_response"]})
-    contexts = relevant_contexts_chain.invoke({"context_number": relevant_response["context_number"], "context": context})
-    final_response = response_chain.invoke({"query": query, "context": contexts["relevant_contexts"]})
+    # Fix: Ensuring all required variables are passed
+    context_relevancy_chain = LLMChain(
+        llm=llm_judge,
+        prompt=PromptTemplate(
+            input_variables=["retriever_query", "context"], 
+            template=relevancy_prompt
+        ),
+        output_key="relevancy_response"
+    )
+
+    relevant_context_chain = LLMChain(
+        llm=llm_judge,
+        prompt=PromptTemplate(
+            input_variables=["relevancy_response"],
+            template=relevant_context_picker_prompt
+        ),
+        output_key="context_number"
+    )
+
+    relevant_contexts_chain = LLMChain(
+        llm=llm_judge,
+        prompt=PromptTemplate(
+            input_variables=["context_number", "context"],
+            template=response_synth
+        ),
+        output_key="relevant_contexts"
+    )
+
+    response_chain = LLMChain(
+        llm=rag_llm,
+        prompt=PromptTemplate(
+            input_variables=["query", "context"],
+            template=rag_prompt
+        ),
+        output_key="final_response"
+    )
+
+    # ----------------- Fix: Ensuring All Keys Exist -----------------
+
+    response_crisis = context_relevancy_chain.invoke({
+        "context": context_str,
+        "retriever_query": query
+    })
+
+    # Debugging: Show intermediate response
+    st.write("🔍 Context Relevancy Response:", response_crisis["relevancy_response"])
+
+    relevant_response = relevant_context_chain.invoke({
+        "relevancy_response": response_crisis["relevancy_response"]
+    })
+
+    st.write("🔍 Picked Relevant Contexts:", relevant_response["context_number"])
+
+    contexts = relevant_contexts_chain.invoke({
+        "context_number": relevant_response["context_number"],
+        "context": context_str  # Ensure correct format
+    })
+
+    st.write("🔍 Extracted Relevant Contexts:", contexts["relevant_contexts"])
+
+    final_response = response_chain.invoke({
+        "query": query,
+        "context": contexts["relevant_contexts"]
+    })
 
     # ----------------- Display All Outputs -----------------
     st.markdown("### Context Relevancy Evaluation")
@@ -175,13 +246,4 @@ if query:
     st.json(contexts["relevant_contexts"])
 
     st.subheader("context_relevancy_evaluation_chain Statement")
-    st.json(final_response["relevancy_response"])
-
-    st.subheader("pick_relevant_context_chain Statement")
-    st.json(final_response["context_number"])
-
-    st.subheader("relevant_contexts_chain Statement")
-    st.json(final_response["relevant_contexts"])
-
-    st.subheader("RAG Response Statement")
     st.json(final_response["final_response"])