ByteHub Cloud
ByteHub Cloud works in exactly the same way as the open-source feature store, except that any features you create are stored on a managed cloud account. This allows you to easily store timeseries features for your models without having to worry about setting up database, cloud storage buckets etc.
In Python, install using:
pip install bytehub[cloud]
Then login to the feature store by running:
fs = CloudFeatureStore()
You'll be prompted to login and paste an authentication code into Python. At this point you will have a CloudFeatureStore object that allows you to create, transform and load features in exactly the same way as the other tutorials. All of your data will be saved into a private AWS S3 storage bucket, allowing you to share and reuse features easily between sessions.
Last modified 2yr ago