Class Activations

Activations can either be used through an Activation layer, or through the activation argument supported by all forward layers:

Inheritance

System.Object

Keras

Base

Activations

Implements

System.IDisposable

Inherited Members

Base.Parameters

Base.None

Base.Init()

Base.ToPython()

Base.InvokeStaticMethod(Object, String, Dictionary<String, Object>)

Base.InvokeMethod(String, Dictionary<String, Object>)

Base.Item[String]

Keras.Instance

Keras.keras

Keras.keras2onnx

Keras.tfjs

Keras.Dispose()

Keras.ToTuple(Array)

Keras.ToList(Array)

System.Object.Equals(System.Object)

System.Object.Equals(System.Object, System.Object)

System.Object.GetHashCode()

System.Object.GetType()

System.Object.MemberwiseClone()

System.Object.ReferenceEquals(System.Object, System.Object)

System.Object.ToString()

Namespace: Keras

Assembly: Keras.dll

Syntax

public class Activations : Base, IDisposable

Methods

| Improve this Doc View Source

Elu(NDarray, Single)

Declaration

public static NDarray Elu(NDarray x, float alpha = 1F)

Parameters

Type	Name	Description
Numpy.NDarray	x
System.Single	alpha

Returns

Type	Description
Numpy.NDarray

| Improve this Doc View Source

Exponential(NDarray)

Exponential (base e) activation function.

Declaration

public static NDarray Exponential(NDarray x)

Parameters

Type	Name	Description
Numpy.NDarray	x	Input tensor.

Returns

Type	Description
Numpy.NDarray	Output tensor

| Improve this Doc View Source

HardSigmoid(NDarray)

Hard sigmoid activation function. Faster to compute than sigmoid activation.

Declaration

public static NDarray HardSigmoid(NDarray x)

Parameters

Type	Name	Description
Numpy.NDarray	x	Input tensor.

Returns

Type	Description
Numpy.NDarray	Output tensor

| Improve this Doc View Source

Linear(NDarray)

Linear (i.e. identity) activation function.

Declaration

public static NDarray Linear(NDarray x)

Parameters

Type	Name	Description
Numpy.NDarray	x	Input tensor.

Returns

Type	Description
Numpy.NDarray	Output tensor

| Improve this Doc View Source

Relu(NDarray, Single, Nullable<Single>, Single)

Declaration

public static NDarray Relu(NDarray x, float alpha = 0F, float? max_value = default(float? ), float threshold = 0F)

Parameters

Type	Name	Description
Numpy.NDarray	x
System.Single	alpha
System.Nullable<System.Single>	max_value
System.Single	threshold

Returns

Type	Description
Numpy.NDarray

| Improve this Doc View Source

Selu(NDarray)

Scaled Exponential Linear Unit (SELU). SELU is equal to: scale* elu(x, alpha), where alpha and scale are predefined constants.The values of alpha and scale are chosen so that the mean and variance of the inputs are preserved between two consecutive layers as long as the weights are initialized correctly(see lecun_normal initialization) and the number of inputs is "large enough" (see references for more information).

Declaration

public static NDarray Selu(NDarray x)

Parameters

Type	Name	Description
Numpy.NDarray	x	Input tensor.

Returns

Type	Description
Numpy.NDarray	The scaled exponential unit activation: scale * elu(x, alpha).

| Improve this Doc View Source

Sigmoid(NDarray)

Sigmoid activation function.

Declaration

public static NDarray Sigmoid(NDarray x)

Parameters

Type	Name	Description
Numpy.NDarray	x	Input tensor.

Returns

Type	Description
Numpy.NDarray	Tensor, output of sigmoid

| Improve this Doc View Source

Softmax(NDarray, Int32)

Softmax activation function

Declaration

public static NDarray Softmax(NDarray x, int axis = -1)

Parameters

Type	Name	Description
Numpy.NDarray	x	The input tensor.
System.Int32	axis	Integer, axis along which the softmax normalization is applied.

Returns

Type	Description
Numpy.NDarray	Tensor, output of softmax transformation.

| Improve this Doc View Source

Softplus(NDarray)

Softplus activation function. The softplus activation: log(exp(x) + 1).

Declaration

public static NDarray Softplus(NDarray x)

Parameters

Type	Name	Description
Numpy.NDarray	x	Input tensor.

Returns

Type	Description
Numpy.NDarray	Output tensor

| Improve this Doc View Source

Softsign(NDarray)

The softsign activation: x / (abs(x) + 1).

Declaration

public static NDarray Softsign(NDarray x)

Parameters

Type	Name	Description
Numpy.NDarray	x	Input tensor.

Returns

Type	Description
Numpy.NDarray	Output tensor

| Improve this Doc View Source

Tanh(NDarray)

Hyperbolic tangent activation function.

Declaration

public static NDarray Tanh(NDarray x)

Parameters

Type	Name	Description
Numpy.NDarray	x	Input tensor.

Returns

Type	Description
Numpy.NDarray	Output tensor

Implements

System.IDisposable