Update app.py

app.py

@@ -28,16 +28,15 @@ if "vector_store" not in st.session_state:
     st.session_state.vector_store = None
 if "documents" not in st.session_state:
     st.session_state.documents = None
-if "processed" not in st.session_state:
-    st.session_state.processed = False  # Prevent redundant processing
+if "pdf_loaded" not in st.session_state:
+    st.session_state.pdf_loaded = False
+if "chunked" not in st.session_state:
+    st.session_state.chunked = False
+if "vector_created" not in st.session_state:
+    st.session_state.vector_created = False
 
-# Step 1: Choose PDF Source (Horizontal radio buttons)
-pdf_source = st.radio(
-    "Upload or provide a link to a PDF:", 
-    ["Upload a PDF file", "Enter a PDF URL"], 
-    index=0, 
-    horizontal=True
-)
+# Step 1: Choose PDF Source
+pdf_source = st.radio("Upload or provide a link to a PDF:", ["Upload a PDF file", "Enter a PDF URL"], index=0, horizontal=True)
 
 pdf_path = None
 if pdf_source == "Upload a PDF file":
@@ -47,7 +46,9 @@ if pdf_source == "Upload a PDF file":
         with open(pdf_path, "wb") as f:
             f.write(uploaded_file.getbuffer())
         st.success("✅ PDF Uploaded Successfully!")
-        st.session_state.processed = False  # Reset processing
+        st.session_state.pdf_loaded = False  
+        st.session_state.chunked = False
+        st.session_state.vector_created = False
 
 elif pdf_source == "Enter a PDF URL":
     pdf_url = st.text_input("Enter PDF URL:")
@@ -60,44 +61,50 @@ elif pdf_source == "Enter a PDF URL":
                     with open(pdf_path, "wb") as f:
                         f.write(response.content)
                     st.success("✅ PDF Downloaded Successfully!")
-                    st.session_state.processed = False  # Reset processing
+                    st.session_state.pdf_loaded = False
+                    st.session_state.chunked = False
+                    st.session_state.vector_created = False
                 else:
                     st.error("❌ Failed to download PDF. Check the URL.")
             except Exception as e:  
                 st.error(f"Error downloading PDF: {e}")
 
-
-# Step 2: Process PDF and Create Vector Store (Only if Not Processed)
-if pdf_path and not st.session_state.processed:
-    with st.spinner("Loading and processing PDF..."):
+# Step 2: Process PDF
+if pdf_path and not st.session_state.pdf_loaded:
+    with st.spinner("Loading PDF..."):
         loader = PDFPlumberLoader(pdf_path)
         docs = loader.load()
+        st.session_state.documents = docs
+        st.session_state.pdf_loaded = True  
         st.success(f"✅ **PDF Loaded!** Total Pages: {len(docs)}")
 
-        # Step 3: Chunking
-        with st.spinner("Chunking the document..."):
-            model_name = "nomic-ai/modernbert-embed-base"
-            embedding_model = HuggingFaceEmbeddings(model_name=model_name, model_kwargs={'device': 'cpu'}, encode_kwargs={'normalize_embeddings': False})
-            text_splitter = SemanticChunker(embedding_model)
-            documents = text_splitter.split_documents(docs)
-            st.session_state.documents = documents
-            st.success(f"✅ **Document Chunked!** Total Chunks: {len(documents)}")
-
-        # Step 4: Setup Vectorstore
-        with st.spinner("Creating vector store..."):
-            vector_store = Chroma(
-                collection_name="deepseek_collection",
-                collection_metadata={"hnsw:space": "cosine"},
-                embedding_function=embedding_model
-            )
-            vector_store.add_documents(documents)
-            num_documents = len(vector_store.get()["documents"])
-            st.session_state.vector_store = vector_store  # Store in session state
-            st.session_state.processed = True  # Mark as processed
-            st.success(f"✅ **Vector Store Created!** Total documents stored: {num_documents}")
-
-# Step 5: Query Input (Only allow if vector store exists)
-if st.session_state.vector_store:
+# Step 3: Chunking (Only if Not Already Done)
+if st.session_state.pdf_loaded and not st.session_state.chunked:
+    with st.spinner("Chunking the document..."):
+        model_name = "nomic-ai/modernbert-embed-base"
+        embedding_model = HuggingFaceEmbeddings(model_name=model_name, model_kwargs={'device': 'cpu'}, encode_kwargs={'normalize_embeddings': False})
+        text_splitter = SemanticChunker(embedding_model)
+        documents = text_splitter.split_documents(st.session_state.documents)
+        st.session_state.documents = documents
+        st.session_state.chunked = True  
+        st.success(f"✅ **Document Chunked!** Total Chunks: {len(documents)}")
+
+# Step 4: Setup Vectorstore
+if st.session_state.chunked and not st.session_state.vector_created:
+    with st.spinner("Creating vector store..."):
+        vector_store = Chroma(
+            collection_name="deepseek_collection",
+            collection_metadata={"hnsw:space": "cosine"},
+            embedding_function=embedding_model
+        )
+        vector_store.add_documents(st.session_state.documents)
+        num_documents = len(vector_store.get()["documents"])
+        st.session_state.vector_store = vector_store  
+        st.session_state.vector_created = True  
+        st.success(f"✅ **Vector Store Created!** Total documents stored: {num_documents}")
+
+# Step 5: Query Input
+if st.session_state.vector_created:
     query = st.text_input("🔍 Enter a Query:")
     if query:
         with st.spinner("Retrieving relevant contexts..."):
@@ -145,5 +152,14 @@ if st.session_state.vector_store:
         st.subheader("🟥 RAG Final Response")
         st.success(final_response['final_response'])
 
+        # Final + Intermediate Outputs
+        st.subheader("🔍 **Full Workflow Breakdown:**")
+        st.json({
+            "Context Relevancy Evaluation": relevancy_response["relevancy_response"],
+            "Relevant Contexts": relevant_response["context_number"],
+            "Extracted Contexts": final_contexts["relevant_contexts"],
+            "Final Answer": final_response["final_response"]
+        })
+
 else:
     st.warning("📄 Please upload or provide a PDF URL first.")