Class: Google::Apis::AiplatformV1::GoogleCloudAiplatformV1StudySpec

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

Represents specification of a Study.

Instance Attribute Summary collapse

Instance Method Summary collapse

Constructor Details

#initialize(**args) ⇒ GoogleCloudAiplatformV1StudySpec

Returns a new instance of GoogleCloudAiplatformV1StudySpec.



28593
28594
28595
# File 'lib/google/apis/aiplatform_v1/classes.rb', line 28593

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

Instance Attribute Details

#algorithmString

The search algorithm specified for the Study. Corresponds to the JSON property algorithm

Returns:

  • (String)


28532
28533
28534
# File 'lib/google/apis/aiplatform_v1/classes.rb', line 28532

def algorithm
  @algorithm
end

#convex_automated_stopping_specGoogle::Apis::AiplatformV1::GoogleCloudAiplatformV1StudySpecConvexAutomatedStoppingSpec

Configuration for ConvexAutomatedStoppingSpec. When there are enough completed trials (configured by min_measurement_count), for pending trials with enough measurements and steps, the policy first computes an overestimate of the objective value at max_num_steps according to the slope of the incomplete objective value curve. No prediction can be made if the curve is completely flat. If the overestimation is worse than the best objective value of the completed trials, this pending trial will be early-stopped, but a last measurement will be added to the pending trial with max_num_steps and predicted objective value from the autoregression model. Corresponds to the JSON property convexAutomatedStoppingSpec



28545
28546
28547
# File 'lib/google/apis/aiplatform_v1/classes.rb', line 28545

def convex_automated_stopping_spec
  @convex_automated_stopping_spec
end

#decay_curve_stopping_specGoogle::Apis::AiplatformV1::GoogleCloudAiplatformV1StudySpecDecayCurveAutomatedStoppingSpec

The decay curve automated stopping rule builds a Gaussian Process Regressor to predict the final objective value of a Trial based on the already completed Trials and the intermediate measurements of the current Trial. Early stopping is requested for the current Trial if there is very low probability to exceed the optimal value found so far. Corresponds to the JSON property decayCurveStoppingSpec



28554
28555
28556
# File 'lib/google/apis/aiplatform_v1/classes.rb', line 28554

def decay_curve_stopping_spec
  @decay_curve_stopping_spec
end

#measurement_selection_typeString

Describe which measurement selection type will be used Corresponds to the JSON property measurementSelectionType

Returns:

  • (String)


28559
28560
28561
# File 'lib/google/apis/aiplatform_v1/classes.rb', line 28559

def measurement_selection_type
  @measurement_selection_type
end

#median_automated_stopping_specGoogle::Apis::AiplatformV1::GoogleCloudAiplatformV1StudySpecMedianAutomatedStoppingSpec

The median automated stopping rule stops a pending Trial if the Trial's best objective_value is strictly below the median 'performance' of all completed Trials reported up to the Trial's last measurement. Currently, 'performance' refers to the running average of the objective values reported by the Trial in each measurement. Corresponds to the JSON property medianAutomatedStoppingSpec



28568
28569
28570
# File 'lib/google/apis/aiplatform_v1/classes.rb', line 28568

def median_automated_stopping_spec
  @median_automated_stopping_spec
end

#metricsArray<Google::Apis::AiplatformV1::GoogleCloudAiplatformV1StudySpecMetricSpec>

Required. Metric specs for the Study. Corresponds to the JSON property metrics



28573
28574
28575
# File 'lib/google/apis/aiplatform_v1/classes.rb', line 28573

def metrics
  @metrics
end

#observation_noiseString

The observation noise level of the study. Currently only supported by the Vertex AI Vizier service. Not supported by HyperparameterTuningJob or TrainingPipeline. Corresponds to the JSON property observationNoise

Returns:

  • (String)


28580
28581
28582
# File 'lib/google/apis/aiplatform_v1/classes.rb', line 28580

def observation_noise
  @observation_noise
end

#parametersArray<Google::Apis::AiplatformV1::GoogleCloudAiplatformV1StudySpecParameterSpec>

Required. The set of parameters to tune. Corresponds to the JSON property parameters



28585
28586
28587
# File 'lib/google/apis/aiplatform_v1/classes.rb', line 28585

def parameters
  @parameters
end

#study_stopping_configGoogle::Apis::AiplatformV1::GoogleCloudAiplatformV1StudySpecStudyStoppingConfig

The configuration (stopping conditions) for automated stopping of a Study. Conditions include trial budgets, time budgets, and convergence detection. Corresponds to the JSON property studyStoppingConfig



28591
28592
28593
# File 'lib/google/apis/aiplatform_v1/classes.rb', line 28591

def study_stopping_config
  @study_stopping_config
end

Instance Method Details

#update!(**args) ⇒ Object

Update properties of this object



28598
28599
28600
28601
28602
28603
28604
28605
28606
28607
28608
# File 'lib/google/apis/aiplatform_v1/classes.rb', line 28598

def update!(**args)
  @algorithm = args[:algorithm] if args.key?(:algorithm)
  @convex_automated_stopping_spec = args[:convex_automated_stopping_spec] if args.key?(:convex_automated_stopping_spec)
  @decay_curve_stopping_spec = args[:decay_curve_stopping_spec] if args.key?(:decay_curve_stopping_spec)
  @measurement_selection_type = args[:measurement_selection_type] if args.key?(:measurement_selection_type)
  @median_automated_stopping_spec = args[:median_automated_stopping_spec] if args.key?(:median_automated_stopping_spec)
  @metrics = args[:metrics] if args.key?(:metrics)
  @observation_noise = args[:observation_noise] if args.key?(:observation_noise)
  @parameters = args[:parameters] if args.key?(:parameters)
  @study_stopping_config = args[:study_stopping_config] if args.key?(:study_stopping_config)
end