Class: Google::Apis::AiplatformV1::GoogleCloudAiplatformV1SupervisedTuningSpec

Inherits:
Object
  • Object
show all
Includes:
Core::Hashable, Core::JsonObjectSupport
Defined in:
lib/google/apis/aiplatform_v1/classes.rb,
lib/google/apis/aiplatform_v1/representations.rb,
lib/google/apis/aiplatform_v1/representations.rb

Overview

Tuning Spec for Supervised Tuning for first party models.

Instance Attribute Summary collapse

Instance Method Summary collapse

Constructor Details

#initialize(**args) ⇒ GoogleCloudAiplatformV1SupervisedTuningSpec

Returns a new instance of GoogleCloudAiplatformV1SupervisedTuningSpec.



48279
48280
48281
 
# File 'lib/google/apis/aiplatform_v1/classes.rb', line 48279

def initialize(**args)
   update!(**args)
end

Instance Attribute Details

#evaluation_config ⇒ Google::Apis::AiplatformV1::GoogleCloudAiplatformV1EvaluationConfig

Evaluation Config for Tuning Job. Corresponds to the JSON property evaluationConfig

Returns:



48250
48251
48252
 
# File 'lib/google/apis/aiplatform_v1/classes.rb', line 48250

def evaluation_config
  @evaluation_config
end

#export_last_checkpoint_only ⇒ Boolean Also known as: export_last_checkpoint_only?

Optional. If set to true, disable intermediate checkpoints for SFT and only the last checkpoint will be exported. Otherwise, enable intermediate checkpoints for SFT. Default is false. Corresponds to the JSON property exportLastCheckpointOnly

Returns:

  • (Boolean) 


48257
48258
48259
 
# File 'lib/google/apis/aiplatform_v1/classes.rb', line 48257

def export_last_checkpoint_only
  @export_last_checkpoint_only
end

#hyper_parameters ⇒ Google::Apis::AiplatformV1::GoogleCloudAiplatformV1SupervisedHyperParameters

Hyperparameters for SFT. Corresponds to the JSON property hyperParameters

Returns:



48263
48264
48265
 
# File 'lib/google/apis/aiplatform_v1/classes.rb', line 48263

def hyper_parameters
  @hyper_parameters
end

#training_dataset_uri ⇒ String

Required. Training dataset used for tuning. The dataset can be specified as either a Cloud Storage path to a JSONL file or as the resource name of a Vertex Multimodal Dataset. Corresponds to the JSON property trainingDatasetUri

Returns:

  • (String) 


48270
48271
48272
 
# File 'lib/google/apis/aiplatform_v1/classes.rb', line 48270

def training_dataset_uri
  @training_dataset_uri
end

#validation_dataset_uri ⇒ String

Optional. Validation dataset used for tuning. The dataset can be specified as either a Cloud Storage path to a JSONL file or as the resource name of a Vertex Multimodal Dataset. Corresponds to the JSON property validationDatasetUri

Returns:

  • (String) 


48277
48278
48279
 
# File 'lib/google/apis/aiplatform_v1/classes.rb', line 48277

def validation_dataset_uri
  @validation_dataset_uri
end

Instance Method Details

#update!(**args) ⇒ Object

Update properties of this object



48284
48285
48286
48287
48288
48289
48290
 
# File 'lib/google/apis/aiplatform_v1/classes.rb', line 48284

def update!(**args)
  @evaluation_config = args[:evaluation_config] if args.key?(:evaluation_config)
  @export_last_checkpoint_only = args[:export_last_checkpoint_only] if args.key?(:export_last_checkpoint_only)
  @hyper_parameters = args[:hyper_parameters] if args.key?(:hyper_parameters)
  @training_dataset_uri = args[:training_dataset_uri] if args.key?(:training_dataset_uri)
  @validation_dataset_uri = args[:validation_dataset_uri] if args.key?(:validation_dataset_uri)
end