Update app.py

app.py

@@ -14,8 +14,6 @@ from langchain.vectorstores import Chroma
 from dotenv import load_dotenv, find_dotenv
 _ = load_dotenv(find_dotenv())
 
-global vectordb
-
 #openai.api_key = os.environ["OPENAI_API_KEY"]
 
 template = """Use the following pieces of context to answer the question at the end. If you don't know the answer, just say that you don't know, don't try to make up 
@@ -24,9 +22,12 @@ template = """Use the following pieces of context to answer the question at the
 
 QA_CHAIN_PROMPT = PromptTemplate(input_variables = ["context", "question"], template = template)
 
+print(os.listdir("docs/chroma/"))
+print(os.listdir("docs/youtube/"))
+
 def invoke(openai_api_key, youtube_url, prompt):
     openai.api_key = openai_api_key
-    if vectordb is not None:
+    if (len(os.listdir("docs/chroma/")) == 0):
         youtube_dir = "docs/youtube/"
         loader = GenericLoader(YoutubeAudioLoader([youtube_url], youtube_dir), OpenAIWhisperParser())
         docs = loader.load()
@@ -37,8 +38,8 @@ def invoke(openai_api_key, youtube_url, prompt):
         llm = ChatOpenAI(model_name = "gpt-4", temperature = 0)
         qa_chain = RetrievalQA.from_chain_type(llm, retriever = vectordb.as_retriever(), return_source_documents = True, chain_type_kwargs = {"prompt": QA_CHAIN_PROMPT})
     result = qa_chain({"query": prompt})
-    #shutil.rmtree(youtube_dir)
-    #shutil.rmtree(chroma_dir)
+    shutil.rmtree(youtube_dir)
+    shutil.rmtree(chroma_dir)
     return result["result"]
 
 description = """The app demonstrates how to use a <strong>Large Language Model</strong> (LLM) with <strong>Retrieval Augmented Generation</strong> (RAG) on external data.