Upload app.py

app.py

@@ -16,7 +16,7 @@ from langchain_core.messages import HumanMessage, AIMessage
 # Without this you will need to copy/paste the API key with every change
 try:
     # CHANGE the location of the file
-    load_dotenv('C:\\Users\\raj\\.jupyter\\.env')
+    load_dotenv('C:\\Users\\raj\\.jupyter\\.env1')
     # Add the API key to the session - use it for populating the interface
     if os.getenv('OPENAI_API_KEY'):
         st.session_state['OPENAI_API_KEY'] = os.getenv('OPENAI_API_KEY')
@@ -35,6 +35,12 @@ if 'OPENAI_API_KEY' in st.session_state:
 else:
     openai_api_key = st.sidebar.text_input('OpenAI API key',placeholder='copy & paste your OpenAI API key')
 
+if len(openai_api_key) == 0 :
+    "provide valid OpenAI API Key !!"
+    st.stop()
+
+# print(type(openai_api_key))
+# st.stop()
 ### 2. Define utility functions to invoke the LLM
 
 # Create an instance of the LLM for summarization
@@ -83,7 +89,7 @@ def  get_llm_response(prompt):
         return response
 
 # Initialize the session state memory
-if 'MEMORY' not in st.session_state:
+if 'MEMORY' not in st.session_state :
     memory = ConversationSummaryMemory(
         llm = get_summarization_llm(),
         human_prefix='user',
@@ -98,35 +104,37 @@ if 'MEMORY' not in st.session_state:
 # This is needed as streamlit re-runs the entire script when user provides input in a widget
 # https://docs.streamlit.io/develop/api-reference/chat/st.chat_message
 
-for msg in st.session_state['MEMORY'].chat_memory.messages:
 
-    if (isinstance(msg, HumanMessage)):
-        st.chat_message('user').write(msg.content)
-    elif (isinstance(msg, AIMessage)):
-        st.chat_message('ai').write(msg.content)
-    else:
-        print('System message: ', msg.content)
-    
+if  openai_api_key is not None:
+    for msg in st.session_state['MEMORY'].chat_memory.messages:
+
+        if (isinstance(msg, HumanMessage)):
+            st.chat_message('user').write(msg.content)
+        elif (isinstance(msg, AIMessage)):
+            st.chat_message('ai').write(msg.content)
+        else:
+            print('System message: ', msg.content)
+        
 
-### 4. Create the *chat_input* element to get the user query
-# Interface for user input
-prompt = st.chat_input(placeholder='Your input here')
+    ### 4. Create the *chat_input* element to get the user query
+    # Interface for user input
+    prompt = st.chat_input(placeholder='Your input here')
 
-### 5. Process the query received from user
-if prompt and openai_api_key:
+    ### 5. Process the query received from user
+    if prompt and openai_api_key:
 
-    # Write the user prompt as chat message
-    st.chat_message('user').write(prompt)
+        # Write the user prompt as chat message
+        st.chat_message('user').write(prompt)
 
-    # Invoke the LLM
-    response = get_llm_response(prompt)
+        # Invoke the LLM
+        response = get_llm_response(prompt)
 
-    # Write the response as chat_message
-    st.chat_message('ai').write(response['response'])
+        # Write the response as chat_message
+        st.chat_message('ai').write(response['response'])
 
-### 6. Write out the current content of the context
-st.divider()
-st.subheader('Context/Summary:')
+    ### 6. Write out the current content of the context
+    st.divider()
+    st.subheader('Context/Summary:')
 
-# Print the state of the buffer
-st.session_state['MEMORY'].buffer
+    # Print the state of the buffer
+    st.session_state['MEMORY'].buffer