Output only. Error of
[feature_attributions][google.cloud.aiplatform.v1beta1.Attribution.feature_attributions]
caused by approximation used in the explanation method. Lower value means
more precise attributions.
For Sampled Shapley
[attribution][google.cloud.aiplatform.v1beta1.ExplanationParameters.sampled_shapley_attribution],
increasing
[path_count][google.cloud.aiplatform.v1beta1.SampledShapleyAttribution.path_count]
might reduce the error.
For Integrated Gradients
[attribution][google.cloud.aiplatform.v1beta1.ExplanationParameters.integrated_gradients_attribution],
increasing
[step_count][google.cloud.aiplatform.v1beta1.IntegratedGradientsAttribution.step_count]
might reduce the error.
For [XRAI
attribution][google.cloud.aiplatform.v1beta1.ExplanationParameters.xrai_attribution],
increasing
[step_count][google.cloud.aiplatform.v1beta1.XraiAttribution.step_count]
might reduce the error.
Output only. Model predicted output if the input instance is constructed
from the baselines of all the features defined in
[ExplanationMetadata.inputs][google.cloud.aiplatform.v1beta1.ExplanationMetadata.inputs].
The field name of the output is determined by the key in
[ExplanationMetadata.outputs][google.cloud.aiplatform.v1beta1.ExplanationMetadata.outputs].
If the Model's predicted output has multiple dimensions (rank > 1), this is
the value in the output located by
[output_index][google.cloud.aiplatform.v1beta1.Attribution.output_index].
If there are multiple baselines, their output values are averaged.
Output only. Attributions of each explained feature. Features are extracted
from the [prediction
instances][google.cloud.aiplatform.v1beta1.ExplainRequest.instances]
according to [explanation metadata for
inputs][google.cloud.aiplatform.v1beta1.ExplanationMetadata.inputs].
The value is a struct, whose keys are the name of the feature. The values
are how much the feature in the
[instance][google.cloud.aiplatform.v1beta1.ExplainRequest.instances]
contributed to the predicted result.
The format of the value is determined by the feature's input format:
If the feature is a scalar value, the attribution value is a
[floating number][google.protobuf.Value.number_value].
If the feature is an array of scalar values, the attribution value is
an [array][google.protobuf.Value.list_value].
If the feature is a struct, the attribution value is a
[struct][google.protobuf.Value.struct_value]. The keys in the
attribution value struct are the same as the keys in the feature
struct. The formats of the values in the attribution struct are
determined by the formats of the values in the feature struct.
The
[ExplanationMetadata.feature_attributions_schema_uri][google.cloud.aiplatform.v1beta1.ExplanationMetadata.feature_attributions_schema_uri]
field, pointed to by the
[ExplanationSpec][google.cloud.aiplatform.v1beta1.ExplanationSpec] field of
the
[Endpoint.deployed_models][google.cloud.aiplatform.v1beta1.Endpoint.deployed_models]
object, points to the schema file that describes the features and their
attribution values (if it is populated).
Output only. Model predicted output on the corresponding [explanation
instance][ExplainRequest.instances]. The field name of the output is
determined by the key in
[ExplanationMetadata.outputs][google.cloud.aiplatform.v1beta1.ExplanationMetadata.outputs].
If the Model predicted output has multiple dimensions, this is the value in
the output located by
[output_index][google.cloud.aiplatform.v1beta1.Attribution.output_index].
Output only. The display name of the output identified by
[output_index][google.cloud.aiplatform.v1beta1.Attribution.output_index].
For example, the predicted class name by a multi-classification Model.
This field is only populated iff the Model predicts display names as a
separate field along with the explained output. The predicted display name
must has the same shape of the explained output, and can be located using
output_index.
Output only. The index that locates the explained prediction output.
If the prediction output is a scalar value, output_index is not populated.
If the prediction output has multiple dimensions, the length of the
output_index list is the same as the number of dimensions of the output.
The i-th element in output_index is the element index of the i-th dimension
of the output vector. Indices start from 0.
Output only. Name of the explain output. Specified as the key in
[ExplanationMetadata.outputs][google.cloud.aiplatform.v1beta1.ExplanationMetadata.outputs].
[[["Easy to understand","easyToUnderstand","thumb-up"],["Solved my problem","solvedMyProblem","thumb-up"],["Other","otherUp","thumb-up"]],[["Missing the information I need","missingTheInformationINeed","thumb-down"],["Too complicated / too many steps","tooComplicatedTooManySteps","thumb-down"],["Out of date","outOfDate","thumb-down"],["Samples / code issue","samplesCodeIssue","thumb-down"],["Other","otherDown","thumb-down"]],["Last updated 2025-08-14 UTC."],[[["\u003cp\u003eThe \u003ccode\u003eAttribution\u003c/code\u003e class in the Vertex AI v1beta1 API provides explanations for prediction outputs, detailing how each feature contributes to a model's prediction.\u003c/p\u003e\n"],["\u003cp\u003eThis class implements several interfaces, including \u003ccode\u003eIMessage\u003c/code\u003e, \u003ccode\u003eIEquatable\u003c/code\u003e, \u003ccode\u003eIDeepCloneable\u003c/code\u003e, and \u003ccode\u003eIBufferMessage\u003c/code\u003e, and inherits from \u003ccode\u003eobject\u003c/code\u003e, showing its capabilities for message handling, equality checks, deep cloning, and buffering.\u003c/p\u003e\n"],["\u003cp\u003eKey properties include \u003ccode\u003eApproximationError\u003c/code\u003e (indicating the precision of attributions), \u003ccode\u003eBaselineOutputValue\u003c/code\u003e (the model's predicted output from feature baselines), and \u003ccode\u003eFeatureAttributions\u003c/code\u003e (detailing each feature's contribution to the prediction).\u003c/p\u003e\n"],["\u003cp\u003e\u003ccode\u003eInstanceOutputValue\u003c/code\u003e shows the model's predicted output on a specific instance, while \u003ccode\u003eOutputIndex\u003c/code\u003e helps locate this output within multi-dimensional predictions, and \u003ccode\u003eOutputName\u003c/code\u003e describes the output name.\u003c/p\u003e\n"],["\u003cp\u003eConstructors are provided to create new instances of the \u003ccode\u003eAttribution\u003c/code\u003e object, including one that clones another \u003ccode\u003eAttribution\u003c/code\u003e object.\u003c/p\u003e\n"]]],[],null,["# Vertex AI v1beta1 API - Class Attribution (1.0.0-beta47)\n\nVersion latestkeyboard_arrow_down\n\n- [1.0.0-beta47 (latest)](/dotnet/docs/reference/Google.Cloud.AIPlatform.V1Beta1/latest/Google.Cloud.AIPlatform.V1Beta1.Attribution)\n- [1.0.0-beta46](/dotnet/docs/reference/Google.Cloud.AIPlatform.V1Beta1/1.0.0-beta46/Google.Cloud.AIPlatform.V1Beta1.Attribution) \n\n public sealed class Attribution : IMessage\u003cAttribution\u003e, IEquatable\u003cAttribution\u003e, IDeepCloneable\u003cAttribution\u003e, IBufferMessage, IMessage\n\nReference documentation and code samples for the Vertex AI v1beta1 API class Attribution.\n\nAttribution that explains a particular prediction output. \n\nInheritance\n-----------\n\n[object](https://learn.microsoft.com/dotnet/api/system.object) \\\u003e Attribution \n\nImplements\n----------\n\n[IMessage](https://cloud.google.com/dotnet/docs/reference/Google.Protobuf/latest/Google.Protobuf.IMessage-1.html)[Attribution](/dotnet/docs/reference/Google.Cloud.AIPlatform.V1Beta1/latest/Google.Cloud.AIPlatform.V1Beta1.Attribution), [IEquatable](https://learn.microsoft.com/dotnet/api/system.iequatable-1)[Attribution](/dotnet/docs/reference/Google.Cloud.AIPlatform.V1Beta1/latest/Google.Cloud.AIPlatform.V1Beta1.Attribution), [IDeepCloneable](https://cloud.google.com/dotnet/docs/reference/Google.Protobuf/latest/Google.Protobuf.IDeepCloneable-1.html)[Attribution](/dotnet/docs/reference/Google.Cloud.AIPlatform.V1Beta1/latest/Google.Cloud.AIPlatform.V1Beta1.Attribution), [IBufferMessage](https://cloud.google.com/dotnet/docs/reference/Google.Protobuf/latest/Google.Protobuf.IBufferMessage.html), [IMessage](https://cloud.google.com/dotnet/docs/reference/Google.Protobuf/latest/Google.Protobuf.IMessage.html) \n\nInherited Members\n-----------------\n\n[object.GetHashCode()](https://learn.microsoft.com/dotnet/api/system.object.gethashcode) \n[object.GetType()](https://learn.microsoft.com/dotnet/api/system.object.gettype) \n[object.ToString()](https://learn.microsoft.com/dotnet/api/system.object.tostring)\n\nNamespace\n---------\n\n[Google.Cloud.AIPlatform.V1Beta1](/dotnet/docs/reference/Google.Cloud.AIPlatform.V1Beta1/latest/Google.Cloud.AIPlatform.V1Beta1)\n\nAssembly\n--------\n\nGoogle.Cloud.AIPlatform.V1Beta1.dll\n\nConstructors\n------------\n\n### Attribution()\n\n public Attribution()\n\n### Attribution(Attribution)\n\n public Attribution(Attribution other)\n\nProperties\n----------\n\n### ApproximationError\n\n public double ApproximationError { get; set; }\n\nOutput only. Error of\n\\[feature_attributions\\]\\[google.cloud.aiplatform.v1beta1.Attribution.feature_attributions\\]\ncaused by approximation used in the explanation method. Lower value means\nmore precise attributions.\n\n- For Sampled Shapley \\[attribution\\]\\[google.cloud.aiplatform.v1beta1.ExplanationParameters.sampled_shapley_attribution\\], increasing \\[path_count\\]\\[google.cloud.aiplatform.v1beta1.SampledShapleyAttribution.path_count\\] might reduce the error.\n- For Integrated Gradients \\[attribution\\]\\[google.cloud.aiplatform.v1beta1.ExplanationParameters.integrated_gradients_attribution\\], increasing \\[step_count\\]\\[google.cloud.aiplatform.v1beta1.IntegratedGradientsAttribution.step_count\\] might reduce the error.\n- For \\[XRAI attribution\\]\\[google.cloud.aiplatform.v1beta1.ExplanationParameters.xrai_attribution\\], increasing \\[step_count\\]\\[google.cloud.aiplatform.v1beta1.XraiAttribution.step_count\\] might reduce the error.\n\nSee [this introduction](/vertex-ai/docs/explainable-ai/overview)\nfor more information.\n\n### BaselineOutputValue\n\n public double BaselineOutputValue { get; set; }\n\nOutput only. Model predicted output if the input instance is constructed\nfrom the baselines of all the features defined in\n\\[ExplanationMetadata.inputs\\]\\[google.cloud.aiplatform.v1beta1.ExplanationMetadata.inputs\\].\nThe field name of the output is determined by the key in\n\\[ExplanationMetadata.outputs\\]\\[google.cloud.aiplatform.v1beta1.ExplanationMetadata.outputs\\].\n\nIf the Model's predicted output has multiple dimensions (rank \\\u003e 1), this is\nthe value in the output located by\n\\[output_index\\]\\[google.cloud.aiplatform.v1beta1.Attribution.output_index\\].\n\nIf there are multiple baselines, their output values are averaged.\n\n### FeatureAttributions\n\n public Value FeatureAttributions { get; set; }\n\nOutput only. Attributions of each explained feature. Features are extracted\nfrom the \\[prediction\ninstances\\]\\[google.cloud.aiplatform.v1beta1.ExplainRequest.instances\\]\naccording to \\[explanation metadata for\ninputs\\]\\[google.cloud.aiplatform.v1beta1.ExplanationMetadata.inputs\\].\n\nThe value is a struct, whose keys are the name of the feature. The values\nare how much the feature in the\n\\[instance\\]\\[google.cloud.aiplatform.v1beta1.ExplainRequest.instances\\]\ncontributed to the predicted result.\n\nThe format of the value is determined by the feature's input format:\n\n- If the feature is a scalar value, the attribution value is a\n \\[floating number\\]\\[google.protobuf.Value.number_value\\].\n\n- If the feature is an array of scalar values, the attribution value is\n an \\[array\\]\\[google.protobuf.Value.list_value\\].\n\n- If the feature is a struct, the attribution value is a\n \\[struct\\]\\[google.protobuf.Value.struct_value\\]. The keys in the\n attribution value struct are the same as the keys in the feature\n struct. The formats of the values in the attribution struct are\n determined by the formats of the values in the feature struct.\n\nThe\n\\[ExplanationMetadata.feature_attributions_schema_uri\\]\\[google.cloud.aiplatform.v1beta1.ExplanationMetadata.feature_attributions_schema_uri\\]\nfield, pointed to by the\n\\[ExplanationSpec\\]\\[google.cloud.aiplatform.v1beta1.ExplanationSpec\\] field of\nthe\n\\[Endpoint.deployed_models\\]\\[google.cloud.aiplatform.v1beta1.Endpoint.deployed_models\\]\nobject, points to the schema file that describes the features and their\nattribution values (if it is populated).\n\n### InstanceOutputValue\n\n public double InstanceOutputValue { get; set; }\n\nOutput only. Model predicted output on the corresponding \\[explanation\ninstance\\]\\[ExplainRequest.instances\\]. The field name of the output is\ndetermined by the key in\n\\[ExplanationMetadata.outputs\\]\\[google.cloud.aiplatform.v1beta1.ExplanationMetadata.outputs\\].\n\nIf the Model predicted output has multiple dimensions, this is the value in\nthe output located by\n\\[output_index\\]\\[google.cloud.aiplatform.v1beta1.Attribution.output_index\\].\n\n### OutputDisplayName\n\n public string OutputDisplayName { get; set; }\n\nOutput only. The display name of the output identified by\n\\[output_index\\]\\[google.cloud.aiplatform.v1beta1.Attribution.output_index\\].\nFor example, the predicted class name by a multi-classification Model.\n\nThis field is only populated iff the Model predicts display names as a\nseparate field along with the explained output. The predicted display name\nmust has the same shape of the explained output, and can be located using\noutput_index.\n\n### OutputIndex\n\n public RepeatedField\u003cint\u003e OutputIndex { get; }\n\nOutput only. The index that locates the explained prediction output.\n\nIf the prediction output is a scalar value, output_index is not populated.\nIf the prediction output has multiple dimensions, the length of the\noutput_index list is the same as the number of dimensions of the output.\nThe i-th element in output_index is the element index of the i-th dimension\nof the output vector. Indices start from 0.\n\n### OutputName\n\n public string OutputName { get; set; }\n\nOutput only. Name of the explain output. Specified as the key in\n\\[ExplanationMetadata.outputs\\]\\[google.cloud.aiplatform.v1beta1.ExplanationMetadata.outputs\\]."]]
