Update app.py

app.py

@@ -35,7 +35,7 @@ from langchain.docstore.document import Document
 from langchain.vectorstores import Chroma
 from langchain.text_splitter import RecursiveCharacterTextSplitter
 
-from langchain.chains import VectorDBQA
+from langchain.chains import VectorDBQA, ConversationChain
 
 from langchain.document_loaders import UnstructuredFileLoader, TextLoader
 from langchain import PromptTemplate
@@ -212,7 +212,7 @@ def getRAGChain(customerName, customerDistrict, custDetailsPresent, vectordb,llm
     global memory 
     #memory = ConversationBufferWindowMemory(k=3, memory_key="history", input_key="question", initial_memory=conversation_history)
     #memory = ConversationBufferMemory(k=3, memory_key="history", input_key="query", initial_memory=conversation_history)
-    memory = ConversationBufferMemory(input_key="query")
+    memory = ConversationBufferMemory()
     
     # chain = RetrievalQA.from_chain_type(
     #     llm=getLLMModel(llmID),
@@ -229,7 +229,20 @@ def getRAGChain(customerName, customerDistrict, custDetailsPresent, vectordb,llm
     #             input_key="question"),
     #     }
     # )
-    chain = RetrievalQA.from_chain_type(
+    # chain = RetrievalQA.from_chain_type(
+    #     llm=getLLMModel(llmID),
+    #     chain_type='stuff',
+    #     retriever=getRetriever(vectordb),
+    #     memory=memory,
+    #     #retriever=vectordb.as_retriever(),
+    #     verbose=True,
+    #     chain_type_kwargs={
+    #         "verbose": False,
+    #         "prompt": createPrompt(customerName, customerDistrict, custDetailsPresent),
+    #         "memory": memory
+    #     }
+    # )
+    chain = ConversationChain(
         llm=getLLMModel(llmID),
         chain_type='stuff',
         retriever=getRetriever(vectordb),