Update src/main.py

src/main.py

@@ -65,13 +65,15 @@ def ingest_chunks(chunks):
 #Retreival function to get the data from the database and reply to the user
 def retreival(vector_store, k):
     #Creating LLM
-    # llm = ChatOllama(
-    # base_url='https://thewise-ollama-server.hf.space',
-    # model="codellama")
-    llm = Ollama(
+    llm = ChatOllama(
+    base_url='https://thewise-ollama-server.hf.space',
     model="codellama",
-    base_url="https://thewise-ollama-server.hf.space",
     callback_manager=CallbackManager([StreamingStdOutCallbackHandler()]))
+    
+    # llm = Ollama(
+    # model="codellama",
+    # base_url="https://thewise-ollama-server.hf.space",
+    # callback_manager=CallbackManager([StreamingStdOutCallbackHandler()]))
                      
     # Define the system message template
     #Adding CHAT HISTORY to the System template explicitly because mainly Chat history goes to Condense the Human Question with Backround (Not template), but System template goes straight the LLM Chain