Skip to main content
Version: v3.0.0

Authentication

The Python client authenticates through one of two entry points. Both return a FeatureStoreClient. See Starting the client for the full parameter tables and end-to-end examples.

Platform discovery (login)​

h2o_featurestore.login() is the recommended mechanism. It uses H2O AI Cloud discovery to locate the Feature Store and authorization gateway endpoints and to authenticate you based on your environment. Credentials are resolved in the following order of precedence:

  • a token_provider (an h2o_authn.TokenProvider) that you supply when your application already manages token refresh;
  • an explicit platform_token generated from the CLI & API Access page;
  • the ambient environment — a Notebook Engine or a configured H2O CLI — when no argument is provided.
import h2o_featurestore
client = h2o_featurestore.login()

Explicit OIDC (login_custom)​

h2o_featurestore.login_custom() is for advanced or off-cloud deployments where discovery is not available. You provide the Feature Store endpoint, the authorization gateway endpoint, and the OIDC parameters explicitly. The client uses the supplied OIDC refresh_token (together with the issuer URL and client credentials) to obtain access tokens for each call.

import h2o_featurestore
client = h2o_featurestore.login_custom(
    endpoint="https://featurestore.your-domain.com",
    authz_endpoint="https://authz-gateway.your-domain.com",
    refresh_token="my-secret-refresh-token",
    issuer_url="https://auth.your-domain.com/auth/realms/your-realm",
    client_id="oidc-app-client-id",
    client_secret="oidc-app-secret",
)
Feedback