Bias
dllib
backward
AffineLayer DenoisingAutoEncodeLayer Layer ReLULayer SigmoidLayer SoftmaxLayer
biasShape
ModelShape
biases
Model