[Alpha] Vector Database
Last updated
Last updated
Warning: This is an experimental feature. To our knowledge, this is stable, but there are still rough edges in the experience. Contributions are welcome!
Vector database allows user to store and retrieve embeddings. Feast provides general APIs to store and retrieve embeddings.
Below are supported vector databases and implemented features:
Vector Database | Retrieval | Indexing |
---|---|---|
Note: SQLite is in limited access and only working on Python 3.10. It will be updated as sqlite_vec progresses.
See https://github.com/feast-dev/feast-workshop/blob/rag/module_4_rag for an example on how to use vector database.
Run the following commands to prepare the embedding dataset:
The output will be stored in data/city_wikipedia_summaries.csv.
Use the feature_tore.yaml file to initialize the feature store. This will use the data as offline store, and Pgvector as online store.
Run the following command in terminal to apply the feature store configuration:
Note that when you run feast apply
you are going to apply the following Feature View that we will use for retrieval later:
Then run the following command in the terminal to materialize the data to the online store:
First create a feature store instance, and use the retrieve_online_documents
API to retrieve the top 5 similar documents to the specified query.
We offer two Online Store options for Vector Databases. PGVector and SQLite.
If you are using pyenv
to manage your Python versions, you can install the SQLite extension with the following command:
And you can the Feast install package via:
Pgvector
[x]
[ ]
Elasticsearch
[x]
[x]
Milvus
[ ]
[ ]
Faiss
[ ]
[ ]
SQLite
[x]
[ ]