update to app one more time

app.py

@@ -42,12 +42,9 @@ HF_TOKEN = os.environ["HF_TOKEN"]
 document_loader = TextLoader("./data/paul_graham_essays.txt")
 documents = document_loader.load()
 
-### 2. CREATE TEXT SPLITTER AND SPLIT DOCUMENTS
 text_splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=30)
 split_documents = text_splitter.split_documents(documents)
 
-### 3. LOAD HUGGINGFACE EMBEDDINGS
-
 hf_embeddings = HuggingFaceEndpointEmbeddings(
     model=HF_EMBED_ENDPOINT,
     task="feature-extraction",
@@ -79,7 +76,6 @@ hf_retriever = vectorstore.as_retriever()
 1. Define a String Template
 2. Create a Prompt Template from the String Template
 """
-### 1. DEFINE STRING TEMPLATE
 RAG_PROMPT_TEMPLATE = """\
 <|start_header_id|>system<|end_header_id|>
 You are a helpful assistant. You answer user questions based on provided context. If you can't answer the question with the provided context, say you don't know.<|eot_id|>
@@ -100,16 +96,14 @@ rag_prompt = PromptTemplate.from_template(RAG_PROMPT_TEMPLATE)
 """
 1. Create a HuggingFaceEndpoint for the LLM
 """
-### 1. CREATE HUGGINGFACE ENDPOINT FOR LLM
 hf_llm = HuggingFaceEndpoint(
     endpoint_url=HF_LLM_ENDPOINT,
     max_new_tokens=512,
     top_k=10,
     top_p=0.95,
-    typical_p=0.95,
-    temperature=0.01,
-    repetition_penalty=1.03,
-    huggingfacehub_api_token=HF_TOKEN
+    temperature=0.3,
+    repetition_penalty=1.15,
+    huggingfacehub_api_token=HF_TOKEN,
 )
 
 @cl.author_rename
@@ -134,10 +128,9 @@ async def start_chat():
     The user session is a dictionary that is unique to each user session, and is stored in the memory of the server.
     """
 
-    ### BUILD LCEL RAG CHAIN THAT ONLY RETURNS TEXT
     lcel_rag_chain = (
         {"context": itemgetter("query") | hf_retriever, "query": itemgetter("query")}
-        | rag_prompt | hf_llm 
+        | rag_prompt | hf_llm
     )
 
     cl.user_session.set("lcel_rag_chain", lcel_rag_chain)
@@ -155,7 +148,7 @@ async def main(message: cl.Message):
 
     msg = cl.Message(content="")
 
-    async for chunk in lcel_rag_chain.astream(
+    for chunk in await cl.make_async(lcel_rag_chain.stream)(
         {"query": message.content},
         config=RunnableConfig(callbacks=[cl.LangchainCallbackHandler()]),
     ):