Class

MPSMatrixBatchNormalizationGradient

A batch normalization gradient kernel that operates on matrices.

Declaration

class MPSMatrixBatchNormalizationGradient : MPSMatrixBinaryKernel

Topics

Initializers

init?(coder: NSCoder, device: MTLDevice)
init(device: MTLDevice)

Instance Properties

var epsilon: Float
var sourceInputFeatureChannels: Int
var sourceNumberOfFeatureVectors: Int

Instance Methods

func copy(with: NSZone?, device: MTLDevice?) -> Self
func encode(to: MTLCommandBuffer, gradientMatrix: MPSMatrix, inputMatrix: MPSMatrix, mean: MPSVector, varianceVector: MPSVector, gammaVector: MPSVector?, betaVector: MPSVector?, resultGradientForDataMatrix: MPSMatrix, resultGradientForGammaVector: MPSVector?, resultGradientForBetaVector: MPSVector?)
func neuronParameterA() -> Float
func neuronParameterB() -> Float
func neuronParameterC() -> Float
func neuronType() -> MPSCNNNeuronType
func setNeuronType(MPSCNNNeuronType, parameterA: Float, parameterB: Float, parameterC: Float)

Relationships

Inherits From

Conforms To

See Also

Matrix Normalization Operations

class MPSMatrixBatchNormalization

A batch normalization kernel that operates on matrices.

Beta Software

This documentation contains preliminary information about an API or technology in development. This information is subject to change, and software implemented according to this documentation should be tested with final operating system software.

Learn more about using Apple's beta software