Update app.py

app.py

@@ -60,22 +60,35 @@ retriever = vectordb.as_retriever(
 
 from langchain.chains import create_retrieval_chain
 from langchain.chains.combine_documents import create_stuff_documents_chain
-from langchain import hub
+#from langchain import hub
+
 
 READER_MODEL="HuggingFaceH4/zephyr-7b-beta"
 #HuggingFaceH4/zephyr-7b-beta
 #qa = ConversationalRetrievalChain.from_llm(llm=READER_MODEL,retriever=retriever,memory=memory)
 #qa = RetrievalQA.from_chain_type(llm=READER_MODEL,chain_type="map_reduce",retriever=retriever,verbose=True)
 
-retrieval_qa_chat_prompt = hub.pull("langchain-ai/retrieval-qa-chat")
-combine_docs_chain = create_stuff_documents_chain(
-    READER_MODEL, retrieval_qa_chat_prompt
+#retrieval_qa_chat_prompt = hub.pull("langchain-ai/retrieval-qa-chat")
+
+qa_chat_prompt = ChatPromptTemplate.from_template("""Answer the following question based only on the provided context:
+
+<context>
+{Tou are a doctor}
+</context>
+
+Question: {input}""")
+
+docs_chain = create_stuff_documents_chain(
+    READER_MODEL, qa_chat_prompt
 )
-qa = create_retrieval_chain(retriever, combine_docs_chain)
+retrieval_chain = create_retrieval_chain(retriever, docs_chain)
+response = retrieval_chain.invoke({"input": "how can I reverse diabetes?"})
+print(response["answer"])
+
 
-result = qa(question)
-import gradio as gr
-gr.load("READER_MODEL").launch()
+#result = qa(question)
+#import gradio as gr
+#gr.load("READER_MODEL").launch()
 
 #result = ({"query": question})
-print("qa")
+#print("qa")