Reference documentation and code samples for the Google Cloud Ai Platform V1 Client class ExplanationParameters.
Parameters to configure explaining for Model's predictions.
Generated from protobuf message google.cloud.aiplatform.v1.ExplanationParameters
Methods
__construct
Constructor.
| Parameters | |
|---|---|
| Name | Description |
data |
array
Optional. Data for populating the Message object. |
↳ sampled_shapley_attribution |
Google\Cloud\AIPlatform\V1\SampledShapleyAttribution
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. |
↳ integrated_gradients_attribution |
Google\Cloud\AIPlatform\V1\IntegratedGradientsAttribution
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 |
↳ xrai_attribution |
Google\Cloud\AIPlatform\V1\XraiAttribution
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. |
↳ top_k |
int
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. |
↳ output_indices |
Google\Protobuf\ListValue
If populated, only returns attributions that have 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 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). |
getSampledShapleyAttribution
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.
| Returns | |
|---|---|
| Type | Description |
Google\Cloud\AIPlatform\V1\SampledShapleyAttribution|null | |
hasSampledShapleyAttribution
setSampledShapleyAttribution
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.
| Parameter | |
|---|---|
| Name | Description |
var |
Google\Cloud\AIPlatform\V1\SampledShapleyAttribution
|
| Returns | |
|---|---|
| Type | Description |
$this | |
getIntegratedGradientsAttribution
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
| Returns | |
|---|---|
| Type | Description |
Google\Cloud\AIPlatform\V1\IntegratedGradientsAttribution|null | |
hasIntegratedGradientsAttribution
setIntegratedGradientsAttribution
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
| Parameter | |
|---|---|
| Name | Description |
var |
Google\Cloud\AIPlatform\V1\IntegratedGradientsAttribution
|
| Returns | |
|---|---|
| Type | Description |
$this | |
getXraiAttribution
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.
| Returns | |
|---|---|
| Type | Description |
Google\Cloud\AIPlatform\V1\XraiAttribution|null | |
hasXraiAttribution
setXraiAttribution
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.
| Parameter | |
|---|---|
| Name | Description |
var |
Google\Cloud\AIPlatform\V1\XraiAttribution
|
| Returns | |
|---|---|
| Type | Description |
$this | |
getTopK
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.
| Returns | |
|---|---|
| Type | Description |
int | |
setTopK
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.
| Parameter | |
|---|---|
| Name | Description |
var |
int
|
| Returns | |
|---|---|
| Type | Description |
$this | |
getOutputIndices
If populated, only returns attributions that have 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 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).
| Returns | |
|---|---|
| Type | Description |
Google\Protobuf\ListValue|null | |
hasOutputIndices
clearOutputIndices
setOutputIndices
If populated, only returns attributions that have 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 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).
| Parameter | |
|---|---|
| Name | Description |
var |
Google\Protobuf\ListValue
|
| Returns | |
|---|---|
| Type | Description |
$this | |
getMethod
| Returns | |
|---|---|
| Type | Description |
string | |