Cloud AI Platform v1 API - Class ExplanationParameters (3.11.0) 

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

Reference documentation and code samples for the Cloud AI Platform v1 API class ExplanationParameters.

Parameters to configure explaining for Model's predictions.

Inheritance

object > ExplanationParameters

Namespace

Google.Cloud.AIPlatform.V1

Assembly

Google.Cloud.AIPlatform.V1.dll

Constructors

ExplanationParameters()

public ExplanationParameters()

ExplanationParameters(ExplanationParameters)

public ExplanationParameters(ExplanationParameters other)
Parameter
Name Description
other ExplanationParameters

Properties

Examples

public Examples Examples { get; set; }

Example-based explanations that returns the nearest neighbors from the provided dataset.

Property Value
Type Description
Examples

IntegratedGradientsAttribution

public IntegratedGradientsAttribution IntegratedGradientsAttribution { get; set; }

An attribution method that computes Aumann-Shapley values taking advantage of the model's fully differentiable structure. Refer to this paper for more details: https://arxiv.org/abs/1703.01365

Property Value
Type Description
IntegratedGradientsAttribution

MethodCase

public ExplanationParameters.MethodOneofCase MethodCase { get; }
Property Value
Type Description
ExplanationParametersMethodOneofCase

OutputIndices

public ListValue OutputIndices { get; set; }

If populated, only returns attributions that have [output_index][google.cloud.aiplatform.v1.Attribution.output_index] contained in output_indices. It must be an ndarray of integers, with the same shape of the output it's explaining.

If not populated, returns attributions for [top_k][google.cloud.aiplatform.v1.ExplanationParameters.top_k] indices of outputs. If neither top_k nor output_indices is populated, returns the argmax index of the outputs.

Only applicable to Models that predict multiple outputs (e,g, multi-class Models that predict multiple classes).

Property Value
Type Description
ListValue

SampledShapleyAttribution

public SampledShapleyAttribution SampledShapleyAttribution { get; set; }

An attribution method that approximates Shapley values for features that contribute to the label being predicted. A sampling strategy is used to approximate the value rather than considering all subsets of features. Refer to this paper for model details: https://arxiv.org/abs/1306.4265.

Property Value
Type Description
SampledShapleyAttribution

TopK

public int TopK { get; set; }

If populated, returns attributions for top K indices of outputs (defaults to 1). Only applies to Models that predicts more than one outputs (e,g, multi-class Models). When set to -1, returns explanations for all outputs.

Property Value
Type Description
int

XraiAttribution

public XraiAttribution XraiAttribution { get; set; }

An attribution method that redistributes Integrated Gradients attribution to segmented regions, taking advantage of the model's fully differentiable structure. Refer to this paper for more details: https://arxiv.org/abs/1906.02825

XRAI currently performs better on natural images, like a picture of a house or an animal. If the images are taken in artificial environments, like a lab or manufacturing line, or from diagnostic equipment, like x-rays or quality-control cameras, use Integrated Gradients instead.

Property Value
Type Description
XraiAttribution