huberLoss(reductionType:delta:weights:)
Creates a huber loss layer with the reduction type, delta, and weights you specify.
Declaration
class func huberLoss(reductionType: MLCReductionType, delta: Float, weights: MLCTensor?) -> SelfParameters
- reductionType:
The reduction operation type.
- delta:
The delta value.
- weights:
The loss label weights tensor.
Return Value
A huber loss layer.
See Also
Creating Loss Layers with Tensor Weights
softmaxCrossEntropy(reductionType:labelSmoothing:classCount:weights:)categoricalCrossEntropy(reductionType:labelSmoothing:classCount:weights:)sigmoidCrossEntropy(reductionType:labelSmoothing:weights:)log(reductionType:epsilon:weights:)meanAbsoluteError(reductionType:weights:)meanSquaredError(reductionType:weights:)hingeLoss(reductionType:weights:)cosineDistance(reductionType:weights:)