Cloud AI Platform v1beta1 API - Class ModelExplanation (1.0.0-beta05) 

public sealed class ModelExplanation : IMessage<ModelExplanation>, IEquatable<ModelExplanation>, IDeepCloneable<ModelExplanation>, IBufferMessage, IMessage

Reference documentation and code samples for the Cloud AI Platform v1beta1 API class ModelExplanation.

Aggregated explanation metrics for a Model over a set of instances.

Inheritance

object > ModelExplanation

Namespace

Google.Cloud.AIPlatform.V1Beta1

Assembly

Google.Cloud.AIPlatform.V1Beta1.dll

Constructors

ModelExplanation()

public ModelExplanation()

ModelExplanation(ModelExplanation)

public ModelExplanation(ModelExplanation other)
Parameter
Name Description
other ModelExplanation

Properties

MeanAttributions

public RepeatedField<Attribution> MeanAttributions { get; }

Output only. Aggregated attributions explaining the Model's prediction outputs over the set of instances. The attributions are grouped by outputs.

For Models that predict only one output, such as regression Models that predict only one score, there is only one attibution that explains the predicted output. For Models that predict multiple outputs, such as multiclass Models that predict multiple classes, each element explains one specific item. [Attribution.output_index][google.cloud.aiplatform.v1beta1.Attribution.output_index] can be used to identify which output this attribution is explaining.

The [baselineOutputValue][google.cloud.aiplatform.v1beta1.Attribution.baseline_output_value], [instanceOutputValue][google.cloud.aiplatform.v1beta1.Attribution.instance_output_value] and [featureAttributions][google.cloud.aiplatform.v1beta1.Attribution.feature_attributions] fields are averaged over the test data.

NOTE: Currently AutoML tabular classification Models produce only one attribution, which averages attributions over all the classes it predicts. [Attribution.approximation_error][google.cloud.aiplatform.v1beta1.Attribution.approximation_error] is not populated.

Property Value
Type Description
RepeatedFieldAttribution