flytekitplugins.awssagemaker.DistributedTrainingContext

class flytekitplugins.awssagemaker.DistributedTrainingContext(current_host: 'str', hosts: 'typing.List[str]', network_interface_name: 'str')

Methods

Parameters

• current_host (str) –

• hosts (List[str]) –

• network_interface_name (str) –

Return type

None

classmethod from_env()

SageMaker suggests “Hostname information might not be immediately available to the processing container. We recommend adding a retry policy on hostname resolution operations as nodes become available in the cluster.” https://docs.aws.amazon.com/sagemaker/latest/dg/build-your-own-processing-container.html#byoc-config This is why we have an automatic retry policy

Return type

flytekitplugins.awssagemaker.distributed_training.DistributedTrainingContext

classmethod from_sagemaker_context_file()

Return type

flytekitplugins.awssagemaker.distributed_training.DistributedTrainingContext

classmethod local_execute()

Creates a dummy local execution context for distributed execution. TODO revisit if this is a good idea

Return type

flytekitplugins.awssagemaker.distributed_training.DistributedTrainingContext

Attributes

current_host: str

hosts: List[str]

network_interface_name: str