Residual block for the WaveNet TCN — layer

This is a block composed i.a. from causal convolutional layers. It may be considered as a replacement for the recurrent layers.

Usage

layer_tcn(
  object,
  nb_filters = 64,
  kernel_size = 3,
  nb_stacks = 1,
  dilations = c(1, 7, 14),
  padding = "causal",
  use_skip_connections = TRUE,
  dropout_rate = 0,
  return_sequences = FALSE,
  activation = "relu",
  kernel_initializer = "he_normal",
  use_batch_norm = FALSE,
  use_layer_norm = FALSE,
  use_weight_norm = FALSE,
  input_shape = NULL,
  ...
)

Arguments

nb_filters: The number of convolutional filters to use in this block
kernel_size: The size of the convolutional kernel
padding: The padding used in the convolutional layers, 'same' or 'causal'.
dropout_rate: Float between 0 and 1. Fraction of the input units to drop.
activation: The final activation used in o = Activation(x + F(x))
kernel_initializer: Initializer for the kernel weights matrix (Conv1D).
use_batch_norm: Whether to use batch normalization in the residual layers or not.
use_layer_norm: Whether to use layer normalization in the residual layers or not.
use_weight_norm: Whether to use weight normalization in the residual layers or not.
dilation_rate: The dilation power of 2 we are using for this residual block

References

Keras TCN library by Philippe Rémy
Sh. Bai., J.Z Kolter, V. Koltun, An Empirical Evaluation of Generic Convolutional and Recurrent Networks for Sequence Modeling](https://arxiv.org/abs/1803.01271) (2018)

Examples

# \donttest{
inp <- layer_input(c(28, 3))
tcn <- layer_tcn()(inp)
model <- keras_model(inp, tcn)
model(array(1, c(32, 28, 3)))
#> Error in py_call_impl(callable, dots$args, dots$keywords): tensorflow.python.framework.errors_impl.UnimplementedError: Exception encountered when calling layer "conv1D_0" "                 f"(type Conv1D).
#> 
#> {{function_node __wrapped__Conv2D_device_/job:localhost/replica:0/task:0/device:GPU:0}} DNN library is not found. [Op:Conv2D]
#> 
#> Call arguments received by layer "conv1D_0" "                 f"(type Conv1D):
#>   • inputs=tf.Tensor(shape=(32, 28, 3), dtype=float32)
# }