computeNormBackward(input:output:axes:outputGradient:generatingInputGradient:)
Backpropogates gradients for the compute norm function.
Declaration
static func computeNormBackward(input: BNNSNDArrayDescriptor, output: BNNSNDArrayDescriptor, axes: [Int]? = nil, outputGradient: BNNSNDArrayDescriptor, generatingInputGradient inputGradient: BNNSNDArrayDescriptor) throwsParameters
- input:
The descriptor of the input.
- output:
The descriptor of the output.
- axes:
The indices of the axes over which the function computes the norm. Set to
nilto specify that the function computes the norm over the entire tensor. - outputGradient:
The descriptor of the output delta.
- inputGradient:
The descriptor of the input delta.