crested.tl.zoo.utils.dilated_residual

crested.tl.zoo.utils.dilated_residual#

crested.tl.zoo.utils.dilated_residual(inputs, filters, kernel_size=3, rate_mult=2.0, dropout=0, conv_type='standard', repeat=1, round=False, **kwargs)#

Construct a residual dilated convolution block.

Parameters:

  • inputs (KerasTensor) – Input tensor.

  • filters (int) – Number of filters in the convolutional layer.

  • kernel_size (int (default: 3)) – Size of the convolutional kernel.

  • rate_mult (float (default: 2.0)) – Rate multiplier for dilated convolution.

  • dropout (float (default: 0)) – Dropout rate probability.

  • conv_type (str (default: 'standard')) – Conv1D layer type.

  • repeat (int (default: 1)) – Number of times to repeat the block.

  • round (bool (default: False)) – Whether to round the dilation rate.

  • **kwargs – Additional keyword arguments.

Return type:

KerasTensor

Returns:

Output tensor after applying the dilated residual block.

Contents