WarmStartConfig: WarmStartConfig Class

In DyfanJones/sagemaker-r-mlcore: sagemaker machine learning core classes and methods

WarmStartConfig Class

Description

Warm Start Configuration which defines the nature of the warm start “HyperparameterTuner“, with type and parents for warm start. Examples: >>> warm_start_config = WarmStartConfig( >>> type=WarmStartTypes.TransferLearning, parents="p1","p2") >>> warm_start_config.type "TransferLearning" >>> warm_start_config.parents "p1","p2"

Public fields

type

Supported warm start types

parents

Set of parent tuning jobs

Methods

Public methods

• WarmStartConfig$new()

• WarmStartConfig$from_job_desc()

• WarmStartConfig$to_input_req()

• WarmStartConfig$format()

• WarmStartConfig$clone()

---

Method new()

Initializes the “WarmStartConfig“ with the provided “WarmStartTypes“ and parents.

Usage

WarmStartConfig$new(
  warm_start_type = c("IdenticalDataAndAlgorithm", "TransferLearning"),
  parents = NULL
)

Arguments

warm_start_type

(str): This should be one of the supported warm start types “'("IdenticalDataAndAlgorithm", "TransferLearning")“'

parents

(str/list): Set of parent tuning jobs which will be used to warm start the new tuning job.

---

Method from_job_desc()

Creates an instance of “WarmStartConfig“ class, from warm start configuration response from DescribeTrainingJob. Examples: >>> warm_start_config = WarmStartConfig$new()$from_job_desc(warm_start_config=list( >>> "WarmStartType"="TransferLearning", >>> "ParentHyperParameterTuningJobs"= list( >>> list('HyperParameterTuningJobName'= "p1"), >>> list('HyperParameterTuningJobName'= "p2") >>> ) >>>)) >>> warm_start_config.type "TransferLearning" >>> warm_start_config.parents ["p1","p2"]

Usage

WarmStartConfig$from_job_desc(warm_start_config)

Arguments

warm_start_config

(dict): The expected format of the “warm_start_config“ contains two first-class

Returns

sagemaker.tuner.WarmStartConfig: De-serialized instance of WarmStartConfig containing the type and parents provided as part of “warm_start_config“.

---

Method to_input_req()

Converts the “self“ instance to the desired input request format. Examples: >>> warm_start_config = WarmStartConfig$new("TransferLearning",parents="p1,p2") >>> warm_start_config$to_input_req() list( "WarmStartType"="TransferLearning", "ParentHyperParameterTuningJobs"= list( list('HyperParameterTuningJobName': "p1"), list('HyperParameterTuningJobName': "p2") ) )

Usage

WarmStartConfig$to_input_req()

Returns

list: Containing the "WarmStartType" and "ParentHyperParameterTuningJobs" as the first class fields.

---

Method format()

format class

Usage

WarmStartConfig$format()

---

Method clone()

The objects of this class are cloneable with this method.

Usage

WarmStartConfig$clone(deep = FALSE)

Arguments

deep

Whether to make a deep clone.

DyfanJones/sagemaker-r-mlcore documentation built on May 3, 2022, 10:08 a.m.