Update rag.py

rag.py

@@ -22,8 +22,8 @@ WEB_URL       = "https://openai.com/research/gpt-4"
 YOUTUBE_URL_1 = "https://www.youtube.com/watch?v=--khbXchTeE"
 YOUTUBE_URL_2 = "https://www.youtube.com/watch?v=hdhZwyf24mE"
 
-YOUTUBE_DIR = "/data/yt"
-CHROMA_DIR  = "/data/db"
+YOUTUBE_DIR = "/data/youtube"
+CHROMA_DIR  = "/data/chroma"
 
 MONGODB_ATLAS_CLUSTER_URI = os.environ["MONGODB_ATLAS_CLUSTER_URI"]
 MONGODB_DB_NAME           = "langchain_db"
@@ -63,23 +63,23 @@ def document_splitting(config, docs):
     
 def document_storage_chroma(chunks):
     Chroma.from_documents(documents = chunks, 
-                          embedding = OpenAIEmbeddings(), 
+                          embedding = OpenAIEmbeddings(disallowed_special = ()), 
                           persist_directory = CHROMA_DIR)
 
 def document_storage_mongodb(chunks):
     MongoDBAtlasVectorSearch.from_documents(documents = chunks,
-                                            embedding = OpenAIEmbeddings(),
+                                            embedding = OpenAIEmbeddings(disallowed_special = ()),
                                             collection = collection,
                                             index_name = MONGODB_INDEX_NAME)
 
 def document_retrieval_chroma():
-    return Chroma(embedding_function = OpenAIEmbeddings(),
+    return Chroma(embedding_function = OpenAIEmbeddings(disallowed_special = ()),
                   persist_directory = CHROMA_DIR)
 
 def document_retrieval_mongodb():
     return MongoDBAtlasVectorSearch.from_connection_string(MONGODB_ATLAS_CLUSTER_URI,
                                                            MONGODB_DB_NAME + "." + MONGODB_COLLECTION_NAME,
-                                                           OpenAIEmbeddings(),
+                                                           OpenAIEmbeddings(disallowed_special = ()),
                                                            index_name = MONGODB_INDEX_NAME)
 
 def rag_batch(config):