Starting April 29, 2025, Gemini 1.5 Pro and Gemini 1.5 Flash models are not available in projects that have no prior usage of these models, including new projects. For details, see Model versions and lifecycle.
This is similar to
noise_sigma,
but provides additional flexibility. A separate noise sigma can be
provided for each feature, which is useful if their distributions are
different. No noise is added to features that are not set. If this field
is unset,
noise_sigma
will be used for all features.
This is similar to
noise_sigma,
but provides additional flexibility. A separate noise sigma can be
provided for each feature, which is useful if their distributions are
different. No noise is added to features that are not set. If this field
is unset,
noise_sigma
will be used for all features.
This is a single float value and will be used to add noise to all the
features. Use this field when all features are normalized to have the
same distribution: scale to range [0, 1], [-1, 1] or z-scoring, where
features are normalized to have 0-mean and 1-variance. Learn more about
normalization.
For best results the recommended value is about 10% - 20% of the standard
deviation of the input feature. Refer to section 3.2 of the SmoothGrad
paper: https://arxiv.org/pdf/1706.03825.pdf. Defaults to 0.1.
If the distribution is different per feature, set
feature_noise_sigma
instead for each feature.
The number of gradient samples to use for
approximation. The higher this number, the more accurate the gradient
is, but the runtime complexity increases by this factor as well.
Valid range of its value is [1, 50]. Defaults to 3.
This is similar to
noise_sigma,
but provides additional flexibility. A separate noise sigma can be
provided for each feature, which is useful if their distributions are
different. No noise is added to features that are not set. If this field
is unset,
noise_sigma
will be used for all features.
This is a single float value and will be used to add noise to all the
features. Use this field when all features are normalized to have the
same distribution: scale to range [0, 1], [-1, 1] or z-scoring, where
features are normalized to have 0-mean and 1-variance. Learn more about
normalization.
For best results the recommended value is about 10% - 20% of the standard
deviation of the input feature. Refer to section 3.2 of the SmoothGrad
paper: https://arxiv.org/pdf/1706.03825.pdf. Defaults to 0.1.
If the distribution is different per feature, set
feature_noise_sigma
instead for each feature.
[[["Easy to understand","easyToUnderstand","thumb-up"],["Solved my problem","solvedMyProblem","thumb-up"],["Other","otherUp","thumb-up"]],[["Hard to understand","hardToUnderstand","thumb-down"],["Incorrect information or sample code","incorrectInformationOrSampleCode","thumb-down"],["Missing the information/samples I need","missingTheInformationSamplesINeed","thumb-down"],["Other","otherDown","thumb-down"]],["Last updated 2025-09-03 UTC."],[],[],null,["# Interface SmoothGradConfigOrBuilder (1.32.0)\n\n public interface SmoothGradConfigOrBuilder extends MessageOrBuilder\n\nImplements\n----------\n\n[MessageOrBuilder](https://cloud.google.com/java/docs/reference/protobuf/latest/com.google.protobuf.MessageOrBuilder.html)\n\nMethods\n-------\n\n### getFeatureNoiseSigma()\n\n public abstract FeatureNoiseSigma getFeatureNoiseSigma()\n\nThis is similar to\nnoise_sigma,\nbut provides additional flexibility. A separate noise sigma can be\nprovided for each feature, which is useful if their distributions are\ndifferent. No noise is added to features that are not set. If this field\nis unset,\nnoise_sigma\nwill be used for all features.\n\n`.google.cloud.vertexai.v1.FeatureNoiseSigma feature_noise_sigma = 2;`\n\n### getFeatureNoiseSigmaOrBuilder()\n\n public abstract FeatureNoiseSigmaOrBuilder getFeatureNoiseSigmaOrBuilder()\n\nThis is similar to\nnoise_sigma,\nbut provides additional flexibility. A separate noise sigma can be\nprovided for each feature, which is useful if their distributions are\ndifferent. No noise is added to features that are not set. If this field\nis unset,\nnoise_sigma\nwill be used for all features.\n\n`.google.cloud.vertexai.v1.FeatureNoiseSigma feature_noise_sigma = 2;`\n\n### getGradientNoiseSigmaCase()\n\n public abstract SmoothGradConfig.GradientNoiseSigmaCase getGradientNoiseSigmaCase()\n\n### getNoiseSigma()\n\n public abstract float getNoiseSigma()\n\nThis is a single float value and will be used to add noise to all the\nfeatures. Use this field when all features are normalized to have the\nsame distribution: scale to range \\[0, 1\\], \\[-1, 1\\] or z-scoring, where\nfeatures are normalized to have 0-mean and 1-variance. Learn more about\n[normalization](https://developers.google.com/machine-learning/data-prep/transform/normalization).\n\nFor best results the recommended value is about 10% - 20% of the standard\ndeviation of the input feature. Refer to section 3.2 of the SmoothGrad\npaper: \u003chttps://arxiv.org/pdf/1706.03825.pdf\u003e. Defaults to 0.1.\n\nIf the distribution is different per feature, set\nfeature_noise_sigma\ninstead for each feature.\n\n`float noise_sigma = 1;`\n\n### getNoisySampleCount()\n\n public abstract int getNoisySampleCount()\n\nThe number of gradient samples to use for\napproximation. The higher this number, the more accurate the gradient\nis, but the runtime complexity increases by this factor as well.\nValid range of its value is \\[1, 50\\]. Defaults to 3.\n\n`int32 noisy_sample_count = 3;`\n\n### hasFeatureNoiseSigma()\n\n public abstract boolean hasFeatureNoiseSigma()\n\nThis is similar to\nnoise_sigma,\nbut provides additional flexibility. A separate noise sigma can be\nprovided for each feature, which is useful if their distributions are\ndifferent. No noise is added to features that are not set. If this field\nis unset,\nnoise_sigma\nwill be used for all features.\n\n`.google.cloud.vertexai.v1.FeatureNoiseSigma feature_noise_sigma = 2;`\n\n### hasNoiseSigma()\n\n public abstract boolean hasNoiseSigma()\n\nThis is a single float value and will be used to add noise to all the\nfeatures. Use this field when all features are normalized to have the\nsame distribution: scale to range \\[0, 1\\], \\[-1, 1\\] or z-scoring, where\nfeatures are normalized to have 0-mean and 1-variance. Learn more about\n[normalization](https://developers.google.com/machine-learning/data-prep/transform/normalization).\n\nFor best results the recommended value is about 10% - 20% of the standard\ndeviation of the input feature. Refer to section 3.2 of the SmoothGrad\npaper: \u003chttps://arxiv.org/pdf/1706.03825.pdf\u003e. Defaults to 0.1.\n\nIf the distribution is different per feature, set\nfeature_noise_sigma\ninstead for each feature.\n\n`float noise_sigma = 1;`"]]
