Update app.py

app.py

@@ -41,19 +41,9 @@ def get_vector_store_from_url(url):
     return vector_store
     
 
-def get_context_retriever_chain(vector_store):
+def get_context_retriever_chain(vector_store,llm):
     # llm = ChatOpenAI()
-    llm = CTransformers(
-                    # model = "TheBloke/Mistral-7B-Instruct-v0.2-GGUF",
-                    model= "TheBloke/Llama-2-7B-Chat-GGUF",
-                    model_file = "llama-2-7b-chat.Q3_K_S.gguf",
-                    model_type="llama",
-                    max_new_tokens = 300,
-                    temperature = 0.3,
-                    lib="avx2", # for CPU
-                    )
-
-    
+    llm = llm
     retriever = vector_store.as_retriever()
     
     prompt = ChatPromptTemplate.from_messages([
@@ -67,17 +57,9 @@ def get_context_retriever_chain(vector_store):
     return retriever_chain
     
 
-def get_conversational_rag_chain(retriever_chain): 
+def get_conversational_rag_chain(retriever_chain,llm): 
     
-    llm = CTransformers(
-                    # model = "TheBloke/Mistral-7B-Instruct-v0.2-GGUF",
-                    model= "TheBloke/Llama-2-7B-Chat-GGUF",
-                    model_file = "llama-2-7b-chat.Q3_K_S.gguf",
-                    model_type="llama",
-                    max_new_tokens = 300,
-                    temperature = 0.3,
-                    lib="avx2", # for CPU
-    )
+    llm=llm
     
     prompt = ChatPromptTemplate.from_messages([
       ("system", "Answer the user's questions based on the below context:\n\n{context}"),
@@ -90,8 +72,17 @@ def get_conversational_rag_chain(retriever_chain):
     return create_retrieval_chain(retriever_chain, stuff_documents_chain)
 
 def get_response(user_input):
-    retriever_chain = get_context_retriever_chain(st.session_state.vector_store)
-    conversation_rag_chain = get_conversational_rag_chain(retriever_chain)
+    llm = CTransformers(
+                    # model = "TheBloke/Mistral-7B-Instruct-v0.2-GGUF",
+                    model= "TheBloke/Llama-2-7B-Chat-GGUF",
+                    model_file = "llama-2-7b-chat.Q3_K_S.gguf",
+                    model_type="llama",
+                    max_new_tokens = 300,
+                    temperature = 0.3,
+                    lib="avx2", # for CPU
+    )
+    retriever_chain = get_context_retriever_chain(st.session_state.vector_store,llm)
+    conversation_rag_chain = get_conversational_rag_chain(retriever_chain,llm)
     
     response = conversation_rag_chain.invoke({
         "chat_history": st.session_state.chat_history,