Update app.py

app.py

@@ -19,7 +19,6 @@ async def create_youtube_transcription(youtube_url: str) -> List[langchain_core.
     Returns:
         List[langchain_core.documents.Document]: A list of documents containing the youtube transcription
     """
-    await cl.Message(content=f"Hi").send()
     try:
         loader = YoutubeLoader.from_youtube_url(
             youtube_url, add_video_info=False
@@ -59,7 +58,7 @@ async def create_faiss_vector_store(docs: List[langchain_core.documents.Document
     except Exception as e:
         await cl.Message(content=f"failed to create vector db: {e}").send() # display the error if we failed to create the vector db
 
-def create_bm25_retreiver(docs: List[langchain_core.documents.Document]) -> BM25Retriever:
+async def create_bm25_retreiver(docs: List[langchain_core.documents.Document]) -> BM25Retriever:
     """
     Create a BM25 retriever from a list of documents
     More Info: https://python.langchain.com/docs/integrations/retrievers/bm25/
@@ -68,11 +67,14 @@ def create_bm25_retreiver(docs: List[langchain_core.documents.Document]) -> BM25
     Returns:
         BM25Retriever: A BM25 retriever containing the documents
     """
-    bm25 = BM25Retriever.from_documents(docs) # we don't need embeddings for BM25, as it uses keyword matching!
-    bm25.k = 5 # we set k to 5, so we get 5 documents back
-    return bm25
+    try:
+        bm25 = BM25Retriever.from_documents(docs) # we don't need embeddings for BM25, as it uses keyword matching!
+        bm25.k = 5 # we set k to 5, so we get 5 documents back
+        return bm25
+    except Exception as e:
+        await cl.Message(content=f"failed to create BM25 retreiver: {e}").send() # display the error if we failed to create the BM25 retreiver
 
-def create_ensemble_retriever(vector_db:FAISS, bm25:BM25Retriever) -> EnsembleRetriever:
+async def create_ensemble_retriever(vector_db:FAISS, bm25:BM25Retriever) -> EnsembleRetriever:
     """
     Create an ensemble retriever from a vector db and a BM25 retriever
     More Info: https://python.langchain.com/docs/how_to/ensemble_retriever/
@@ -82,8 +84,11 @@ def create_ensemble_retriever(vector_db:FAISS, bm25:BM25Retriever) -> EnsembleRe
     Returns:
         EnsembleRetriever: An ensemble retriever containing the vector db and the BM25 retriever
     """
-    ensemble_retreiver = EnsembleRetriever(retrievers=[vector_db.as_retriever(), bm25], weights=[.3, .7]) # 30% semantic, 70% keyword retrieval
-    return ensemble_retreiver
+    try:
+        ensemble_retreiver = EnsembleRetriever(retrievers=[vector_db.as_retriever(), bm25], weights=[.3, .7]) # 30% semantic, 70% keyword retrieval
+        return ensemble_retreiver
+    except Exception as e:
+        await cl.Message(content=f"failed to create ensemble retriever: {e}").send() # display the error if we failed to create the ensemble retriever
 
 @cl.on_chat_start
 async def start():
@@ -110,8 +115,8 @@ async def start():
         await cl.Message(content=f"youtube docs: {transcription}").send() # display the transcription of the first document to show that we have the correct data
         split_docs = await create_text_splitter(youtube_docs) # split the documents into chunks
         vector_db = await create_faiss_vector_store(split_docs) # create the vector db
-        bm25 = create_bm25_retreiver(split_docs) # create the BM25 retreiver
-        ensemble_retriever = create_ensemble_retriever(vector_db, bm25) # create the ensemble retriever
+        bm25 = await create_bm25_retreiver(split_docs) # create the BM25 retreiver
+        ensemble_retriever = await create_ensemble_retriever(vector_db, bm25) # create the ensemble retriever
         cl.user_session.set("ensemble_retriever", ensemble_retriever) # store the ensemble retriever in the user session for our on message function
     except Exception as e:
         await cl.Message(content=f"failed to load model: {e}").send() # display the error if we failed to load the model