Update main.py

main.py

@@ -29,14 +29,15 @@ import datetime
 
 os.environ["TOKENIZERS_PARALLELISM"] = os.environ["TOKENIZERS_PARALLELISM"]
 os.environ['ANTHROPIC_API_KEY'] = os.environ['ANTHROPIC_API_KEY']
+index_name = os.environ['PINECONE_INDEX_NAME']
+embeddings = HuggingFaceEmbeddings()
+pinecone.init(
+    api_key=os.environ['PINECONE_API_KEY'],
+    environment=os.environ['PINECONE_ENVIRONMENT']
+)
+    
 @cl.cache
-def retriever_to_cache():
-    index_name = os.environ['PINECONE_INDEX_NAME']
-    embeddings = HuggingFaceEmbeddings()
-    pinecone.init(
-        api_key=os.environ['PINECONE_API_KEY'],
-        environment=os.environ['PINECONE_ENVIRONMENT']
-    )
+def retriever_to_cache(index_name, embeddings):
     vectorstore = Pinecone.from_existing_index(
         index_name=index_name, embedding=embeddings
     )
@@ -169,7 +170,7 @@ async def start():
         )
         #question_generator = LLMChain(llm=llm, prompt=CONDENSE_QUESTION_PROMPT)
         #doc_chain = load_qa_chain(streaming_llm, chain_type="stuff")
-        relevant=retriever_to_cache()
+        relevant=retriever_to_cache(index_name, embeddings)
         qa = ConversationalRetrievalChain.from_llm(
             streaming_llm,
             chain_type="stuff",