Class WavenetResidualModule

numpy_ml/neural_nets/modules/modules.py:119–357

class WavenetResidualModule(ModuleBase):
    def __init__(
        self,
        ch_residual,
        ch_dilation,
        dilation,
        kernel_width,
        optimizer=None,
        init="glorot_uniform",
    ):
        """
        A WaveNet-like residual block with causal dilated convolutions.

        .. code-block:: text

            *Skip path in* >-------------------------------------------> + ---> *Skip path out*
                              Causal      |--> Tanh --|                  |
            *Main    |--> Dilated Conv1D -|           * --> 1x1 Conv1D --|
             path >--|                    |--> Sigm --|                  |
             in*     |-------------------------------------------------> + ---> *Main path out*
                                         *Residual path*

        On the final block, the output of the skip path is further processed to
        produce the network predictions.

        References
        ----------
        .. [1] van den Oord et al. (2016). "Wavenet: a generative model for raw
           audio". https://arxiv.org/pdf/1609.03499.pdf

        Parameters
        ----------
        ch_residual : int
            The number of output channels for the 1x1
            :class:`~numpy_ml.neural_nets.layers.Conv1D` layer in the main path.
        ch_dilation : int
            The number of output channels for the causal dilated
            :class:`~numpy_ml.neural_nets.layers.Conv1D` layer in the main path.
        dilation : int
            The dilation rate for the causal dilated
            :class:`~numpy_ml.neural_nets.layers.Conv1D` layer in the main path.
        kernel_width : int
            The width of the causal dilated
            :class:`~numpy_ml.neural_nets.layers.Conv1D` kernel in the main
            path.
        init : {'glorot_normal', 'glorot_uniform', 'he_normal', 'he_uniform'}
            The weight initialization strategy. Default is 'glorot_uniform'.
        optimizer : str or :doc:`Optimizer <numpy_ml.neural_nets.optimizers>` object or None
            The optimization strategy to use when performing gradient updates
            within the :meth:`update` method.  If None, use the
            :class:`~numpy_ml.neural_nets.optimizers.SGD` optimizer with default
            parameters. Default is None.
        """
        super().__init__()

        self.init = init
        self.dilation = dilation
        self.optimizer = optimizer