Table of Contents  ▸  List of Commands  ▸  Neural Networks  ▸  nn_resconv2dnnl

◀  nn_resconv2dnl    |    nn_resconv3d  ▶

nn_resconv2dnnl

Arguments:

• out,_in,_kernel_size>0,_dilation>0,_boundary_conditions,_activation,0<=_learning_mode<=3,_regularization>=0

Description:

Add a resconv2dnnl (residual 2D convolutional layer followed by a normalization layer, then a non-linearity)to the current network.

boundary_conditions can be { 0:dirichlet | 1:neumann | 2:periodic | 3:mirror }.
learning_mode can be { 0:no learning | 1:weights only | 2:biases only | 3:weights+biases }.

Default values:

in=. (previous layer), kernel_size=3, dilation=1, boundary_conditions=1, activation=leakyrelu, learning_mode=3 and regularization=0.