moar log

app.py

@@ -18,6 +18,8 @@ print("""
  STARTING
 =================================================================================
 """)
+
+# Download the data set and save as CSV if it doesn't exist yet.
 csv_path = "./imdb.csv"
 if not os.path.exists(csv_path):
     dataset = load_dataset("ShubhamChoksi/IMDB_Movies")
@@ -26,9 +28,11 @@ if not os.path.exists(csv_path):
 loader = CSVLoader(file_path=csv_path)
 data = loader.load()
 
+# Split data in chunks
 text_splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=100)
 chunked_documents = text_splitter.split_documents(data)
 
+# Store the chunked documents in the vector store if that was not done already
 embedding_model = OpenAIEmbeddings()
 store = LocalFileStore("./cache/")
 embedder = CacheBackedEmbeddings.from_bytes_store(embedding_model, store, namespace=embedding_model.model)