Upload personalized_ht4.py

personalized_ht4.py

@@ -98,40 +98,49 @@ with col1:
     st.markdown("*Multistate ML Analysis Showcase Hypertension*")
 
 with col2:
-    openai_api_key = st.text_input(
+    
+    if 'openai_api_key' not in st.session_state:
+        st.session_state.openai_api_key = ""
+
+    openai_api_key_input = st.text_input(
         "🔑 OpenAI API Key",
         type="password",
         placeholder="sk-...",
-        help="Enter your OpenAI API key to enable AI features"
+        help="Enter your OpenAI API key to enable AI features",
+        value=st.session_state.openai_api_key
     )
 
+    if openai_api_key_input != st.session_state.openai_api_key:
+        st.session_state.openai_api_key = openai_api_key_input
+
+    openai_api_key = st.session_state.openai_api_key
+
     if openai_api_key:
         st.success("✓ API Key set")
     else:
-        st.warning("⚠️ Enter API key")
-        
+        st.warning("⚠️ Enter API key")    
+           
 
 # ali Check if API key is provided
-def get_llm():
+def get_llm(api_key=None):
     """Initialize LangChain LLM with OpenAI - with rate limiting"""
-    if not openai_api_key:
+    if not api_key:
+        api_key = st.session_state.get('openai_api_key', None)
+    
+    if not api_key:
         return None
 
-    # ✅ API 調用限制（每 session 最多 100 次）
     if st.session_state.api_call_count >= 100:
-        st.error("⚠️ API call limit reached (100 calls per session). Please start a new session.")
+        st.error("⚠️ API call limit reached (100 calls per session).")
         st.stop()
 
     try:
         llm = ChatOpenAI(
             model="gpt-4o-mini",
             temperature=0.7,
-            openai_api_key=openai_api_key
+            openai_api_key=api_key
         )
-        
-        # ✅ 記錄 API 調用
         st.session_state.api_call_count += 1
-        
         return llm
     except Exception as e:
         st.error(f"Error initializing OpenAI: {str(e)}")
@@ -139,16 +148,17 @@ def get_llm():
 
 
 # Create vector store from patient data
-def create_patient_vectorstore(patients_df: pd.DataFrame):
-    """Create vector store from patient dataframe for RAG retrieval - USER ISOLATED"""
-    if not openai_api_key:
+def create_patient_vectorstore(patients_df: pd.DataFrame, api_key: str = None):
+    """Create vector store from patient dataframe for RAG retrieval (memory-only)"""
+    if not api_key:
+        api_key = st.session_state.get('openai_api_key', None)
+    
+    if not api_key:
         return None
 
     try:
-        # ✅ 建立用戶專屬目錄
+        import time
         user_id = st.session_state.user_id
-        persist_dir = f"./data/chroma/{user_id}"
-        os.makedirs(persist_dir, exist_ok=True)
         
         documents = []
         for idx, row in patients_df.iterrows():
@@ -163,61 +173,42 @@ Betel: {row.get('betel', 'No')}, Family History: {row['family_history']}"""
 
             doc = Document(
                 page_content=patient_text,
-                metadata={
-                    "patient_id": row['patient_id'],
-                    "user_id": user_id  # ✅ 加入 user_id 標記
-                }
+                metadata={"patient_id": row['patient_id']}
             )
             documents.append(doc)
 
-        embeddings = OpenAIEmbeddings(openai_api_key=openai_api_key)
+        embeddings = OpenAIEmbeddings(openai_api_key=api_key)
         
-        # ✅ 使用用戶專屬的 collection 和目錄
+        # ✅ 純記憶體模式，避免文件鎖問題
         vectorstore = Chroma(
-            collection_name=f"user_{user_id}_patients",
-            embedding_function=embeddings,
-            persist_directory=persist_dir
+            collection_name=f"patients_{user_id}_{int(time.time())}",
+            embedding_function=embeddings
+            # 不設置 persist_directory = 純記憶體
         )
         
-        # 清空舊資料（如果有）
-        try:
-            vectorstore.delete_collection()
-            vectorstore = Chroma(
-                collection_name=f"user_{user_id}_patients",
-                embedding_function=embeddings,
-                persist_directory=persist_dir
-            )
-        except:
-            pass
-        
-        # 加入新文件
-        vectorstore.add_documents(documents)
-        
+        vectorstore.add_documents(documents)        
         return vectorstore
 
     except Exception as e:
         st.error(f"Error creating vector store: {str(e)}")
         return None
-
-
+   
+   
 # Retrieve patient by ID
 def retrieve_patient_by_id(patient_id: str):
-    """Retrieve patient from USER-SPECIFIC dataframe by ID"""
-    user_id = st.session_state.user_id
-    session_key = f"patients_df_{user_id}"
-    
-    # ✅ 從用戶專屬的 DataFrame 檢索
-    if session_key not in st.session_state or st.session_state[session_key] is None:
+    """Retrieve patient from dataframe by ID"""
+    if st.session_state.patients_df is None:
         return None
 
-    patients_df = st.session_state[session_key]
+    patients_df = st.session_state.patients_df
     patient_row = patients_df[patients_df['patient_id'] == patient_id]
 
     if patient_row.empty:
         return None
 
-    return patient_row.iloc[0].to_dict()
-    
+    return patient_row.iloc[0].to_dict()   
+   
+   
 # Sidebar for patient information
 st.sidebar.header("👤 Patient Information")
 
@@ -244,10 +235,8 @@ with st.sidebar.expander("📂 Upload Patient Database (Optional)", expanded=Fal
             else:
                 df = pd.read_excel(uploaded_file)
 
-            # ✅ 修改：使用用戶專屬的 key
-            user_id = st.session_state.user_id
-            session_key = f"patients_df_{user_id}"
-            st.session_state[session_key] = df
+            # ✅ 修改：使用用戶專屬的 key            
+            st.session_state.patients_df = df
             
             # ✅ 記錄檔案 hash
             file_hash = hashlib.md5(uploaded_file.getvalue()).hexdigest()
@@ -255,51 +244,45 @@ with st.sidebar.expander("📂 Upload Patient Database (Optional)", expanded=Fal
                 st.session_state.uploaded_files.append(file_hash)
             
             st.success(f"✅ Loaded {len(df)} patients ({file_size_mb:.1f}MB)")
-            
-            # Vector store 建立（如果需要）
+                       
+            # Vector store 建立
             if openai_api_key and st.button("🔄 Create Vector Store for Smart Search"):
                 with st.spinner("Creating isolated vector store..."):
-                    vectorstore = create_patient_vectorstore(df)
+                    vectorstore = create_patient_vectorstore(df, st.session_state.openai_api_key)
                     if vectorstore:
                         st.session_state.vectorstore = vectorstore
-                        st.success("✅ Vector store created! (Isolated to your session)")
+                        st.success("✅ Vector store created!")
 
         except Exception as e:
-            st.error(f"Error loading file: {str(e)}")
+            st.error(f"Error loading file: {str(e)}")            
+            
 
     # Patient ID retrieval
     # ✅ 修正：使用用戶專屬的 DataFrame key
-    user_id = st.session_state.user_id
-    session_key = f"patients_df_{user_id}"
-    
-    # 顯示已上傳的病患數量（可選）
-    if session_key in st.session_state and st.session_state[session_key] is not None:
-        df = st.session_state[session_key]
+    if st.session_state.patients_df is not None:
+        df = st.session_state.patients_df
         st.caption(f"📊 {len(df)} patients loaded")
     
-    if session_key in st.session_state and st.session_state[session_key] is not None:
+    if st.session_state.patients_df is not None:
         st.markdown("---")
         patient_id_input = st.text_input("🔍 Enter Patient ID", placeholder="P001")
 
-        if st.button("🔥 Load Patient Data"):
+        if st.button("📥 Load Patient Data"):
             if patient_id_input:
                 patient_data = retrieve_patient_by_id(patient_id_input)
                 if patient_data:
-                    # 儲存病患資料
-                    user_id = st.session_state.user_id
-                    loaded_key = f"loaded_patient_{user_id}"
-                    st.session_state[loaded_key] = patient_data
+                    # ✅ 只用簡單命名
                     st.session_state.loaded_patient = patient_data
                     
-                    # ✅ 清除 Tab 6 的狀態
+                    # 清除 Tab 6 的狀態
                     st.session_state.summary_generated = False
                     st.session_state.recommendation_messages = []
-                    
-                    # ✅ 清除 CEA 結果（可選，因為新病人的分析應該重新做）
                     st.session_state.cea_results = None
                     
                     st.success(f"✅ Loaded {patient_id_input}.")
                     st.rerun()
+                else:
+                    st.error(f"❌ Patient ID '{patient_id_input}' not found.")
 
 st.sidebar.markdown("---")
 
@@ -550,7 +533,7 @@ with tab1:
             with st.chat_message("assistant"):
                 with st.spinner("Thinking..."):
                     try:
-                        llm = get_llm()
+                        llm = get_llm(st.session_state.openai_api_key)
                         if llm:
                             history_text = ""
                             for msg in st.session_state.assistant_messages[-10:]:
@@ -1592,7 +1575,7 @@ with tab6:
             if st.button("✨ Generate Personalized Health Summary", type="primary"):
                 with st.spinner("Analyzing your health profile..."):
                     try:
-                        llm = get_llm()
+                        llm = get_llm(st.session_state.openai_api_key)
                         if llm:
                             patient_info = get_patient_info_string()
                             cea_results = get_cea_results_string()
@@ -1643,7 +1626,7 @@ Format the response with clear sections and bullet points."""
                 with st.chat_message("assistant"):
                     with st.spinner("Thinking..."):
                         try:
-                            llm = get_llm()
+                            llm = get_llm(st.session_state.openai_api_key)
                             if llm:
                                 patient_info = get_patient_info_string()
                                 cea_results = get_cea_results_string()