azure_ml_sdk.services.asset.environment

Classes

EnvironmentManager

Class for handling environments in Azure Machine Learning workspace.

Module Contents

class azure_ml_sdk.services.asset.environment.EnvironmentManager(ml_client: azure.ai.ml.MLClient)

Bases: altametris.azure_ml_sdk.services.asset.base.BaseAssetManager

Class for handling environments in Azure Machine Learning workspace.

static _from_docker_image(docker_image: str, environment_name: str, environment_description: str | None = None, environment_version: str | None = None) → azure.ai.ml.entities.Environment

Create en environment from a Docker image.

Parameters:

  • docker_image (str) – The URI of a docker image hosted in a Docker Hub or Azure Container Registry.

  • environment_name (str) – The name of the environment.

  • environment_description (str) – The description of the environment.

  • environment_version (str) – The version of the environment. Defaults to None.

Returns:

An environment object.

static _from_dockerfile(dockerfile_dir: str, environment_name: str, environment_description: str | None = None, environment_version: str | None = None) → azure.ai.ml.entities.Environment

Create en environment from a Dockerfile.

Parameters:

  • dockerfile_dir (str) – The directory containing the Dockerfile.

  • environment_name (str) – The name of the environment.

  • environment_description (str) – The description of the environment.

  • environment_version (str) – The version of the environment. Defaults to None.

Returns:

An environment object.

static _from_conda_file(docker_image: str, conda_file: str, environment_name: str, environment_description: str | None = None, environment_version: str | None = None) → azure.ai.ml.entities.Environment

Create en environment from a conda specification file.

Parameters:

  • docker_image (str) – The URI of a docker image hosted in a Docker Hub or Azure Container Registry.

  • conda_file (str) – The path to the conda specification file (.yaml).

  • environment_name (str) – The name of the environment.

  • environment_description (str) – The description of the environment.

  • environment_version (str) – The version of the environment. Defaults to None.

Returns:

An environment object.

build_environment(method: str, environment_name: str, environment_description: str | None = None, environment_version: str | None = None, docker_image: str | None = None, dockerfile_dir: str | None = None, conda_file: str | None = None) → azure.ai.ml.entities.Environment

Create and build an environment using a defined method.

Parameters:

  • method (str) – Must be one of ‘docker_image’, ‘dockerfile’, or ‘conda_file’.

  • environment_name (str) – The name of the environment.

  • environment_description (str) – The description of the environment.

  • environment_version (str) – The version of the environment.

  • docker_image (str) – The URI of the Docker image.

  • dockerfile_dir (str) – The directory of the Dockerfile.

  • conda_file (str) – The path to the conda specification file.

Returns:

An environment object.