WorkerConfiguration Class

WorkerConfiguration is the class that holds all the necessary information for the workers to run.

Initialize the WorkerConfiguration

:type azureml.core.runconfig.HistoryConfiguration :param use_gpu: Prameter used to signal whether the default base image should have the packages for

gpu added. This parameter is ignored if environment is set.

yaml file. :type conda_dependencies_file: str

Inheritance

azureml._base_sdk_common.abstract_run_config_element._AbstractRunConfigElement

WorkerConfiguration

Constructor

WorkerConfiguration(node_count, compute_target=None, environment=None, shm_size='2g', history=None, use_gpu=False, pip_packages=None, conda_packages=None, conda_dependencies_file=None, pip_requirements_file=None)

Parameters

node_count

int

Required

Number of worker nodes to be initialized, one worker will run per machine in the compute target.

compute_target

AbstractComputeTarget or str

default value: None

The compute target where the workers will run. This can either be an object or the the compute target's name.

environment

Environment

default value: None

The environment definition for the workers. It includes PythonSection, DockerSection, and environment variables. Any environment option not directly exposed through other parameters to the WorkerConfiguration construction can be set using this parameter. If this parameter is specified, it will be used as a base upon which packages specified in pip_packages and conda_packages will be added.

shm_size

str

default value: 2g

The docker shm_size configuration for the worker.

history

HistoryConfiguration

default value: None

History configuration for the worker's run, this controls which logs folders will be monitored

use_gpu

bool

default value: False

Parameter used to signal whether the default base image should have the packages for gpu added. This parameter is ignored if environment is set.

conda_packages

list

default value: None

A list of strings representing conda packages to be added to the Python environment for the workers.

pip_packages

list

default value: None

A list of strings representing pip packages to be added to the Python environment for the workers

pip_requirements_file

str

default value: None

The relative path to the workers' pip requirements text file. This can be provided in combination with the pip_packages parameter.

conda_dependencies_file

str

default value: None

The relative path to the workers' conda dependencies yaml file.

node_count

int

Required

Number of worker nodes to be initialized, one worker will run per machine in the compute target.

compute_target

<xref:azureml.core.compute_target.ComputeTarget> or str

Required

The compute target where the workers will run. This can either be an object or the the compute target's name.

environment

Environment

Required

The environment definition for the workers. It includes PythonSection, DockerSection, and environment variables. Any environment option not directly exposed through other parameters to the WorkerConfiguration construction can be set using this parameter. If this parameter is specified, it will be used as a base upon which packages specified in pip_packages and conda_packages will be added.

shm_size

str

Required

The docker shm_size configuration for the worker.

history

Required

History configuration for the worker's run, this controls which logs folders will be monitored

conda_packages

list

Required

A list of strings representing conda packages to be added to the Python environment for the workers.

pip_packages

list

Required

A list of strings representing pip packages to be added to the Python environment for the workers

pip_requirements_file

str

Required

The relative path to the workers' pip requirements text file. This can be provided in combination with the pip_packages parameter.

conda_dependencies_file

Required

The relative path to the workers' conda dependencies

Attributes

target

Get the compute target where the worker run is scheduled for execution.

Available cloud compute targets can be found using the function compute_targets

Returns

The target name

Return type

str