flytekitplugins.awssagemaker.HyperparameterTuningJobConfig#

class flytekitplugins.awssagemaker.HyperparameterTuningJobConfig(tuning_strategy, tuning_objective, training_job_early_stopping_type)[source]#

The specification of the hyperparameter tuning process https://docs.aws.amazon.com/sagemaker/latest/dg/automatic-model-tuning-ex-tuning-job.html#automatic-model-tuning-ex-low-tuning-config

Methods

Parameters
classmethod from_flyte_idl(pb2_object)[source]#
Parameters

pb2_object (flyteidl.plugins.sagemaker.hyperparameter_tuning_job_pb2.HyperparameterTuningJobConfig) –

serialize_to_string()#
Return type

str

short_string()#
Return type

Text

to_flyte_idl()[source]#
Return type

flyteidl.plugins.sagemaker.hyperparameter_tuning_job_pb2.HyperparameterTuningJobConfig

verbose_string()#
Return type

Text

Attributes

is_empty#
training_job_early_stopping_type#

Enum value of TrainingJobEarlyStoppingType. When the training jobs launched by the hyperparameter tuning job are not improving significantly, a hyperparameter tuning job can be stopping early. This attribute determines how the early stopping is to be done. Note that there’s only a subset of built-in algorithms that supports early stopping. see: https://docs.aws.amazon.com/sagemaker/latest/dg/automatic-model-tuning-early-stopping.html :rtype: int

tuning_objective#

The target metric and the objective of the hyperparameter tuning. :rtype: HyperparameterTuningObjective

tuning_strategy#

Enum value of HyperparameterTuningStrategy. Setting the strategy used when searching in the hyperparameter space :rtype: int