Update app.py

app.py

@@ -549,7 +549,7 @@ and has {vector_store._collection.count()} chunks.")
 # Can use any of these LLMs for responses, for now I am Gemini-Pro for the bot  (this is for responses now, not embeddings)
 
 
-def instantiate_LLM(LLM_provider,api_key,temperature=0.7,top_p=0.95,model_name=None):
+def instantiate_LLM(LLM_provider,api_key,temperature=0.8,top_p=0.95,model_name=None):
     """Instantiate LLM in Langchain.
     Parameters:
         LLM_provider (str): the LLM provider; in ["OpenAI","Google","HuggingFace"]
@@ -688,13 +688,13 @@ chain = ConversationalRetrievalChain.from_llm(
         template=standalone_question_template),
     combine_docs_chain_kwargs={'prompt': ChatPromptTemplate.from_template(answer_template())},
     condense_question_llm=instantiate_LLM(
-        LLM_provider="Google",api_key=google_api_key,temperature=0.1,
+        LLM_provider="Google",api_key=google_api_key,temperature=0.3,
         model_name="gemini-pro"),
     memory=create_memory("gemini-pro"),
     retriever = compression_retriever_HF,
     #retriever = base_retriever_HF,  #base_retriever_HF
     llm=instantiate_LLM(
-        LLM_provider="Google",api_key=google_api_key,temperature=0.7,
+        LLM_provider="Google",api_key=google_api_key,temperature=0.8,
         model_name="gemini-pro"),
     chain_type= "stuff",
     verbose= True,