Depthwise 1D convolution

layer_depthwise_conv_1d(
  object,
  kernel_size,
  strides = 1L,
  padding = "valid",
  depth_multiplier = 1L,
  data_format = NULL,
  dilation_rate = 1L,
  activation = NULL,
  use_bias = TRUE,
  depthwise_initializer = "glorot_uniform",
  bias_initializer = "zeros",
  depthwise_regularizer = NULL,
  bias_regularizer = NULL,
  activity_regularizer = NULL,
  depthwise_constraint = NULL,
  bias_constraint = NULL,
  ...
)

Arguments

object

What to compose the new Layer instance with. Typically a Sequential model or a Tensor (e.g., as returned by layer_input()). The return value depends on object. If object is:

  • missing or NULL, the Layer instance is returned.

  • a Sequential model, the model with an additional layer is returned.

  • a Tensor, the output tensor from layer_instance(object) is returned.

kernel_size

An integer, specifying the height and width of the 1D convolution window. Can be a single integer to specify the same value for all spatial dimensions.

strides

An integer, specifying the strides of the convolution along the height and width. Can be a single integer to specify the same value for all spatial dimensions. Specifying any stride value != 1 is incompatible with specifying any dilation_rate value != 1.

padding

one of 'valid' or 'same' (case-insensitive). "valid" means no padding. "same" results in padding with zeros evenly to the left/right or up/down of the input such that output has the same height/width dimension as the input.

depth_multiplier

The number of depthwise convolution output channels for each input channel. The total number of depthwise convolution output channels will be equal to filters_in * depth_multiplier.

data_format

A string, one of "channels_last" (default) or "channels_first". The ordering of the dimensions in the inputs. channels_last corresponds to inputs with shape (batch_size, height, width, channels) while channels_first corresponds to inputs with shape (batch_size, channels, height, width). It defaults to the image_data_format value found in your Keras config file at ~/.keras/keras.json. If you never set it, then it will be 'channels_last'.

dilation_rate

A single integer, specifying the dilation rate to use for dilated convolution. Currently, specifying any dilation_rate value != 1 is incompatible with specifying any stride value != 1.

activation

Activation function to use. If you don't specify anything, no activation is applied (see ?activation_relu).

use_bias

Boolean, whether the layer uses a bias vector.

depthwise_initializer

Initializer for the depthwise kernel matrix (see initializer_glorot_uniform). If NULL, the default initializer ("glorot_uniform") will be used.

bias_initializer

Initializer for the bias vector (see keras.initializers). If NULL, the default initializer ('zeros') will be used.

depthwise_regularizer

Regularizer function applied to the depthwise kernel matrix (see regularizer_l1()).

bias_regularizer

Regularizer function applied to the bias vector (see regularizer_l1()).

activity_regularizer

Regularizer function applied to the output of the layer (its 'activation') (see regularizer_l1()).

depthwise_constraint

Constraint function applied to the depthwise kernel matrix (see constraint_maxnorm()).

bias_constraint

Constraint function applied to the bias vector (see constraint_maxnorm()).

...

standard layer arguments.

Details

Depthwise convolution is a type of convolution in which each input channel is convolved with a different kernel (called a depthwise kernel). You can understand depthwise convolution as the first step in a depthwise separable convolution.

It is implemented via the following steps:

  • Split the input into individual channels.

  • Convolve each channel with an individual depthwise kernel with depth_multiplier output channels.

  • Concatenate the convolved outputs along the channels axis.

Unlike a regular 1D convolution, depthwise convolution does not mix information across different input channels.

The depth_multiplier argument determines how many filter are applied to one input channel. As such, it controls the amount of output channels that are generated per input channel in the depthwise step.