Class FeatureDropout<T>

Namespace

AiDotNet.Augmentation.Tabular

Assembly

AiDotNet.dll

Randomly masks (sets to zero) features in tabular data.

public class FeatureDropout<T> : TabularAugmenterBase<T>, IAugmentation<T, Matrix<T>>

Type Parameters

T

The numeric type for calculations.

Inheritance

object

AugmentationBase<T, Matrix<T>>

TabularAugmenterBase<T>

FeatureDropout<T>

Implements

IAugmentation<T, Matrix<T>>

Inherited Members

TabularAugmenterBase<T>.GetFeatureCount(Matrix<T>)

TabularAugmenterBase<T>.GetSampleCount(Matrix<T>)

AugmentationBase<T, Matrix<T>>.NumOps

AugmentationBase<T, Matrix<T>>.OnAugmentationApplied

AugmentationBase<T, Matrix<T>>.Name

AugmentationBase<T, Matrix<T>>.Probability

AugmentationBase<T, Matrix<T>>.IsTrainingOnly

AugmentationBase<T, Matrix<T>>.IsEnabled

AugmentationBase<T, Matrix<T>>.Apply(Matrix<T>, AugmentationContext<T>)

AugmentationBase<T, Matrix<T>>.ApplyAugmentation(Matrix<T>, AugmentationContext<T>)

AugmentationBase<T, Matrix<T>>.GetParameters()

AugmentationBase<T, Matrix<T>>.RaiseAugmentationApplied(AugmentationContext<T>, bool)

object.Equals(object)

object.Equals(object, object)

object.GetHashCode()

object.GetType()

object.MemberwiseClone()

object.ReferenceEquals(object, object)

object.ToString()

Remarks

For Beginners: Feature dropout randomly "hides" some features during training by setting them to zero. This forces the model to learn more robust representations that don't rely too heavily on any single feature.

When to use:

• To prevent overfitting to specific features
• To simulate missing data scenarios
• When you want the model to be robust to feature absence

When NOT to use:

• When all features are critical and cannot be missing
• When features have strong interdependencies that break when one is missing

Constructors

FeatureDropout(double, double, int[]?)

Creates a new feature dropout augmentation.

public FeatureDropout(double dropoutRate = 0.1, double probability = 0.5, int[]? featureIndices = null)

Parameters

dropoutRate double

Probability of dropping each feature (default: 0.1).

probability double

Probability of applying this augmentation (default: 0.5).

featureIndices int[]

Optional array of feature indices that can be dropped.

Properties

DropValue

Gets or sets the value to use for dropped features.

public double DropValue { get; set; }

Property Value

double

Remarks

Default: 0.0

Can be set to other values like -1 or the feature mean.

DropoutRate

Gets the probability of dropping each feature.

public double DropoutRate { get; }

Property Value

double

Remarks

Default: 0.1 (10% of features dropped)

Higher values drop more features. Typical range: 0.05 to 0.3

FeatureIndices

Gets or sets the indices of features that can be dropped.

public int[]? FeatureIndices { get; set; }

Property Value

int[]

Remarks

If null, any feature can be dropped.

Methods

ApplyAugmentation(Matrix<T>, AugmentationContext<T>)

Implement this method to perform the actual augmentation.

protected override Matrix<T> ApplyAugmentation(Matrix<T> data, AugmentationContext<T> context)

Parameters

data Matrix<T>

The input data.

context AugmentationContext<T>

The augmentation context.

Returns

Matrix<T>

The augmented data.

GetParameters()

Gets the parameters of this augmentation.

public override IDictionary<string, object> GetParameters()

Returns

IDictionary<string, object>

A dictionary of parameter names to values.