/AWS1/CL_SGMTRAININGSPEC¶
Defines how the algorithm is used for a training job.
CONSTRUCTOR¶
IMPORTING¶
Required arguments:¶
iv_trainingimage TYPE /AWS1/SGMCONTAINERIMAGE /AWS1/SGMCONTAINERIMAGE¶
The Amazon ECR registry path of the Docker image that contains the training algorithm.
it_supportedtrninstancetypes TYPE /AWS1/CL_SGMTRNINSTANCETYPES_W=>TT_TRAININGINSTANCETYPES TT_TRAININGINSTANCETYPES¶
A list of the instance types that this algorithm can use for training.
it_trainingchannels TYPE /AWS1/CL_SGMCHANNELSPEC=>TT_CHANNELSPECIFICATIONS TT_CHANNELSPECIFICATIONS¶
A list of
ChannelSpecificationobjects, which specify the input sources to be used by the algorithm.
Optional arguments:¶
iv_trainingimagedigest TYPE /AWS1/SGMIMAGEDIGEST /AWS1/SGMIMAGEDIGEST¶
An MD5 hash of the training algorithm that identifies the Docker image used for training.
it_supportedhyperparameters TYPE /AWS1/CL_SGMHYPERPARAMETERSPEC=>TT_HYPERPARAMETERSPECS TT_HYPERPARAMETERSPECS¶
A list of the
HyperParameterSpecificationobjects, that define the supported hyperparameters. This is required if the algorithm supports automatic model tuning.>
iv_supportsdistributedtrn TYPE /AWS1/SGMBOOLEAN /AWS1/SGMBOOLEAN¶
Indicates whether the algorithm supports distributed training. If set to false, buyers can't request more than one instance during training.
it_metricdefinitions TYPE /AWS1/CL_SGMMETRICDEFINITION=>TT_METRICDEFINITIONLIST TT_METRICDEFINITIONLIST¶
A list of
MetricDefinitionobjects, which are used for parsing metrics generated by the algorithm.
it_suppedtunjobobjectivemet TYPE /AWS1/CL_SGMHYPPRMTUNJOBOBJIVE=>TT_HYPERPARAMTUNJOBOBJECTIVES TT_HYPERPARAMTUNJOBOBJECTIVES¶
A list of the metrics that the algorithm emits that can be used as the objective metric in a hyperparameter tuning job.
io_additionals3datasource TYPE REF TO /AWS1/CL_SGMADDLS3DATASOURCE /AWS1/CL_SGMADDLS3DATASOURCE¶
The additional data source used during the training job.
Queryable Attributes¶
TrainingImage¶
The Amazon ECR registry path of the Docker image that contains the training algorithm.
Accessible with the following methods¶
| Method | Description |
|---|---|
GET_TRAININGIMAGE() |
Getter for TRAININGIMAGE, with configurable default |
ASK_TRAININGIMAGE() |
Getter for TRAININGIMAGE w/ exceptions if field has no value |
HAS_TRAININGIMAGE() |
Determine if TRAININGIMAGE has a value |
TrainingImageDigest¶
An MD5 hash of the training algorithm that identifies the Docker image used for training.
Accessible with the following methods¶
| Method | Description |
|---|---|
GET_TRAININGIMAGEDIGEST() |
Getter for TRAININGIMAGEDIGEST, with configurable default |
ASK_TRAININGIMAGEDIGEST() |
Getter for TRAININGIMAGEDIGEST w/ exceptions if field has no |
HAS_TRAININGIMAGEDIGEST() |
Determine if TRAININGIMAGEDIGEST has a value |
SupportedHyperParameters¶
A list of the
HyperParameterSpecificationobjects, that define the supported hyperparameters. This is required if the algorithm supports automatic model tuning.>
Accessible with the following methods¶
| Method | Description |
|---|---|
GET_SUPPORTEDHYPERPARAMETERS() |
Getter for SUPPORTEDHYPERPARAMETERS, with configurable defau |
ASK_SUPPORTEDHYPERPARAMETERS() |
Getter for SUPPORTEDHYPERPARAMETERS w/ exceptions if field h |
HAS_SUPPORTEDHYPERPARAMETERS() |
Determine if SUPPORTEDHYPERPARAMETERS has a value |
SupportedTrainingInstanceTypes¶
A list of the instance types that this algorithm can use for training.
Accessible with the following methods¶
| Method | Description |
|---|---|
GET_SUPPORTEDTRNINSTTYPES() |
Getter for SUPPORTEDTRNINSTANCETYPES, with configurable defa |
ASK_SUPPORTEDTRNINSTTYPES() |
Getter for SUPPORTEDTRNINSTANCETYPES w/ exceptions if field |
HAS_SUPPORTEDTRNINSTTYPES() |
Determine if SUPPORTEDTRNINSTANCETYPES has a value |
SupportsDistributedTraining¶
Indicates whether the algorithm supports distributed training. If set to false, buyers can't request more than one instance during training.
Accessible with the following methods¶
| Method | Description |
|---|---|
GET_SUPPORTSDISTRIBUTEDTRN() |
Getter for SUPPORTSDISTRIBUTEDTRAINING, with configurable de |
ASK_SUPPORTSDISTRIBUTEDTRN() |
Getter for SUPPORTSDISTRIBUTEDTRAINING w/ exceptions if fiel |
HAS_SUPPORTSDISTRIBUTEDTRN() |
Determine if SUPPORTSDISTRIBUTEDTRAINING has a value |
MetricDefinitions¶
A list of
MetricDefinitionobjects, which are used for parsing metrics generated by the algorithm.
Accessible with the following methods¶
| Method | Description |
|---|---|
GET_METRICDEFINITIONS() |
Getter for METRICDEFINITIONS, with configurable default |
ASK_METRICDEFINITIONS() |
Getter for METRICDEFINITIONS w/ exceptions if field has no v |
HAS_METRICDEFINITIONS() |
Determine if METRICDEFINITIONS has a value |
TrainingChannels¶
A list of
ChannelSpecificationobjects, which specify the input sources to be used by the algorithm.
Accessible with the following methods¶
| Method | Description |
|---|---|
GET_TRAININGCHANNELS() |
Getter for TRAININGCHANNELS, with configurable default |
ASK_TRAININGCHANNELS() |
Getter for TRAININGCHANNELS w/ exceptions if field has no va |
HAS_TRAININGCHANNELS() |
Determine if TRAININGCHANNELS has a value |
SupportedTuningJobObjectiveMetrics¶
A list of the metrics that the algorithm emits that can be used as the objective metric in a hyperparameter tuning job.
Accessible with the following methods¶
| Method | Description |
|---|---|
GET_SUPPEDTUNJOBOBJECTIVEMET() |
Getter for SUPPEDTUNINGJOBOBJECTIVEMET, with configurable de |
ASK_SUPPEDTUNJOBOBJECTIVEMET() |
Getter for SUPPEDTUNINGJOBOBJECTIVEMET w/ exceptions if fiel |
HAS_SUPPEDTUNJOBOBJECTIVEMET() |
Determine if SUPPEDTUNINGJOBOBJECTIVEMET has a value |
AdditionalS3DataSource¶
The additional data source used during the training job.
Accessible with the following methods¶
| Method | Description |
|---|---|
GET_ADDITIONALS3DATASOURCE() |
Getter for ADDITIONALS3DATASOURCE |