Core

Client

Client(
    address: str,
    username: str | None = None,
    password: str | None = None,
    token_provider: Callable[[], str] | None = None,
    verify: bool | str = True,
    backend_version_override: str | None = None,
    authentication_method: str | None = None,
)

Connect with a Driverless AI server and interact with it.

Parameters:

• address (str) –

  Full URL to the Driverless AI server to connect with.

• username (str | None, default: None ) –

  username to authenticate with the server.

• password (str | None, default: None ) –

  password to authenticate with the server.

• token_provider (Callable[[], str] | None, default: None ) –

  A function that returns a token to authenticate with the server. This precedes username & password authentication.

• verify (bool | str, default: True ) –

  Enable or disable SSL certificate verification when using HTTP to connect to the server. Pass a string path to a CA bundle file to use it for SSL verification. See requests docs for more details.

• backend_version_override (str | None, default: None ) –

  Disables server version detection and use the specified server version. Set to "latest" to use the most recent server backend support.

• authentication_method (str | None, default: None ) –

  The authentication method used to connect to the Driverless AI server as described in link.

Example: Connect with a username and password

client = driverlessai.Client(
    address="http://localhost:12345",
    username="alice",
    password="alice-password",
)

Example: Connect with an OAuth token

• This assumes the Driverless AI server is configured to allow clients to authenticate through tokens.
• Set up an OAuth token provider with a refresh token from the Driverless AI web UI.

  token_provider = driverlessai.token_providers.OAuth2TokenProvider(
      refresh_token="eyJhbGciOiJIUzI1N...",
      client_id="python_client",
      token_endpoint_url="https://keycloak-server/auth/realms/..."
      token_introspection_url="https://keycloak-server/auth/realms/..."
  )
  

• Then use the token provider to authorize and connect to the server.

  client = driverlessai.Client(
      address="https://localhost:12345",
      token_provider=token_provider.ensure_fresh_token
  )
  

Example: Connect with a newer server version

An older Client version will refuse to connect to a newer Driverless AI server version. Even though it is not recommended, you can still force the Client to connect to the server.

client = driverlessai.Client(
    address="http://localhost:12345",
    username="bob",
    password="bob-password",
    backend_version_override="latest",
)

Example: Connect using an alternative authentication method

client = driverlessai.Client(
    address="http://localhost:12345",
    username="alice",
    password="alice-password",
    authentication_method="ldap",
)

admin property

admin: Admin

Perform administrative tasks on the Driverless AI server.

Driverless AI version requirement

Requires Driverless AI server 1.10.3 or higher.

Returns:

• Admin –

   

autodocs property

autodocs: AutoDocs

Interact with AutoDocs in the Driverless AI server.

Beta API

A beta API that is subject to future changes.

Returns:

• AutoDocs –

   

autoviz property

autoviz: AutoViz

Interact with dataset visualizations in the Driverless AI server.

Returns:

• AutoViz –

   

connectors property

connectors: Connectors

Interact with data sources that are enabled in the Driverless AI server.

Returns:

• Connectors –

   

datasets property

datasets: Datasets

Interact with datasets in the Driverless AI server.

Returns:

• Datasets –

   

deployments property

deployments: Deployments

Interact with deployments in the Driverless AI server.

Driverless AI version requirement

Requires Driverless AI server 1.10.6 or higher.

Beta API

A beta API that is subject to future changes.

Returns:

• Deployments –

   

experiments property

experiments: Experiments

Interact with experiments in the Driverless AI server.

Returns:

• Experiments –

   

mli property

mli: MLI

Interact with experiment interpretations in the Driverless AI server.

Returns:

• MLI –

   

model_diagnostics property

model_diagnostics: ModelDiagnostics

Interact with model diagnostics in the Driverless AI server.

Returns:

• ModelDiagnostics –

   

projects property

projects: Projects

Interact with projects in the Driverless AI server.

Returns:

• Projects –

   

recipes property

recipes: Recipes

Interact with recipes in the Driverless AI server.

Returns:

• Recipes –

   

server property

server: Server

Interact with the connected Driverless AI server

Returns:

• Server –

   

is_server_up

is_server_up(
    address: str, timeout: int = 10, verify: bool | str = False
) -> bool

Checks whether a Driverless AI server is up and running.

Parameters:

• address (str) –

  The full URL to the Driverless AI server.

• timeout (int, default: 10 ) –

  The maximum time in seconds to wait for a response from the server.

• verify (bool | str, default: False ) –

  Enable or disable SSL certificate verification when using HTTP to connect to the server. Pass a string path to a CA bundle file to use it for SSL verification. See requests docs for more details.

Returns:

• bool –

  True if the server is up, otherwise False.

Example

is_up = driverlessai.is_server_up("http://localhost:12345")
if is_up:
    print("Driverless AI server is up")

__version__ module-attribute

__version__: str = build_info['version']

Returns package version.

Returns:

• str –