Update app.py

app.py

@@ -41,7 +41,8 @@ from langchain.document_loaders import UnstructuredFileLoader, TextLoader
 from langchain import PromptTemplate
 
 from langchain.chains import RetrievalQA
-from langchain.memory import ConversationBufferWindowMemory
+#from langchain.memory import ConversationBufferWindowMemory
+from langchain.memory import ConversationBufferMemory
 
 from transformers import LlamaTokenizer, AutoTokenizer
 
@@ -119,7 +120,7 @@ def getLLMModel(LLMID):
             model_kwargs={"temperature": 0.2,"max_new_tokens":2500})
         print("Mistral AI LLM Selected")
     else:
-        llm = OpenAI(model_name="gpt-3.5-turbo-instruct",temperature=0.0)
+        llm = OpenAI(model_name="gpt-3.5-turbo-0125",temperature=0.0)
         print("Open AI LLM Selected")
     return llm
 
@@ -208,7 +209,8 @@ def getRAGChain(customerName, customerDistrict, custDetailsPresent, vectordb,llm
     # Retrieve conversation history if available
     #memory = ConversationBufferWindowMemory(k=3, memory_key="history", input_key="question")
     global memory 
-    memory = ConversationBufferWindowMemory(k=3, memory_key="history", input_key="question", initial_memory=conversation_history)
+    #memory = ConversationBufferWindowMemory(k=3, memory_key="history", input_key="question", initial_memory=conversation_history)
+    memory = ConversationBufferMemory(k=3, memory_key="history", input_key="question", initial_memory=conversation_history)
     
     # chain = RetrievalQA.from_chain_type(
     #     llm=getLLMModel(llmID),
@@ -229,6 +231,7 @@ def getRAGChain(customerName, customerDistrict, custDetailsPresent, vectordb,llm
         llm=getLLMModel(llmID),
         chain_type='stuff',
         retriever=getRetriever(vectordb),
+        memory=memory,
         #retriever=vectordb.as_retriever(),
         verbose=False,
         chain_type_kwargs={