Add docstring.

database/utils.py

@@ -26,6 +26,7 @@ def create_collection(
     vector_size: int = 1536,
     distance: Distance = Distance.COSINE,
 ) -> bool:
+    """Create a collection in Qdrant."""
     logger.info(f'Creating collection: "{name}" with vector size: {vector_size}.')
     return client.recreate_collection(
         collection_name=name,
@@ -53,6 +54,7 @@ def upsert(
     collection: str,
     points: List[PointStruct],
 ) -> UpdateResult:
+    """Upsert data points into a Qdrant collection."""
     return client.upsert(collection_name=collection, points=points)
 
 
@@ -88,10 +90,7 @@ def embed_text(
     client: OpenAI, text: Union[str, list], model: str
 ) -> CreateEmbeddingResponse:
     """
-    - Default model (OpenAI): text-embedding-3-small
-    - Max input Tokens: 8191
-    - TikToken model: cl100k_base
-    - Embedding size: 1536
+    Create embeddings using OpenAI API.
     """
     response = client.embeddings.create(input=text, model=model)
     return response

database/vector_database.py

@@ -17,6 +17,7 @@
 
 
 def main(args: argparse.Namespace) -> None:
+    """Main function to create embeddings and vector database."""
     logger.info("Creating embeddings.")
     create_embeddings(
         scraped_dir=args.scraped_dir,
@@ -34,7 +35,7 @@ def main(args: argparse.Namespace) -> None:
     for path in tqdm(data_paths, total=len(data_paths), desc="Creating collections"):
         # Check if this is necessary
         collection_name = path.stem.replace("-", "_")
-        collection_name = collection_name + "_TESTIC"
+        collection_name = collection_name
         points = load_and_process_embeddings(path=path)
 
         create_collection(client=qdrant_client, name=collection_name)