Class DepthwiseConv2D

Depthwise separable 2D convolution.
Depthwise Separable convolutions consists in performing just the first step in a depthwise spatial convolution(which acts on each input channel separately). The depth_multiplier argument controls how many output channels are generated per input channel in the depthwise step.

Inheritance
System.Object
Keras
Base
BaseLayer
DepthwiseConv2D
Implements
System.IDisposable
Inherited Members
System.Object.Equals(System.Object)
System.Object.Equals(System.Object, System.Object)
System.Object.GetHashCode()
System.Object.GetType()
System.Object.MemberwiseClone()
System.Object.ReferenceEquals(System.Object, System.Object)
System.Object.ToString()
Namespace: Keras.Layers
Assembly: Keras.dll
Syntax
public class DepthwiseConv2D : BaseLayer, IDisposable

Constructors

| Improve this Doc View Source

DepthwiseConv2D(Tuple<Int32, Int32>, Tuple<Int32, Int32>, String, Int32, String, String, Boolean, String, String, String, String, String, String, String, Shape)

Initializes a new instance of the DepthwiseConv2D class.

Declaration
public DepthwiseConv2D(Tuple<int, int> kernel_size, Tuple<int, int> strides = null, string padding = "valid", int depth_multiplier = 1, string data_format = "", string activation = "", bool use_bias = true, string depthwise_initializer = "glorot_uniform", string bias_initializer = "zeros", string depthwise_regularizer = "", string bias_regularizer = "", string activity_regularizer = "", string depthwise_constraint = "", string bias_constraint = "", Shape input_shape = null)
Parameters
Type Name Description
System.Tuple<System.Int32, System.Int32> kernel_size

An integer or tuple/list of 2 integers, specifying the height and width of the 2D convolution window. Can be a single integer to specify the same value for all spatial dimensions.
System.Tuple<System.Int32, System.Int32> strides

An integer or tuple/list of 2 integers, 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.
System.String padding

one of "valid" or "same" (case-insensitive).
System.Int32 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.
System.String data_format

A string, one of "channels_last" or "channels_first". The ordering of the dimensions in the inputs. "channels_last" corresponds to inputs with shape (batch, height, width, channels) while "channels_first" corresponds to inputs with shape (batch, 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'.
System.String activation

Activation function to use (see activations). If you don't specify anything, no activation is applied (ie. 'linear' activation: a(x) = x).
System.Boolean use_bias

Boolean, whether the layer uses a bias vector.
System.String depthwise_initializer

Initializer for the depthwise kernel matrix (see initializers).
System.String bias_initializer

Initializer for the bias vector (see initializers).
System.String depthwise_regularizer

Regularizer function applied to the depthwise kernel matrix (see regularizer).
System.String bias_regularizer

Regularizer function applied to the bias vector (see regularizer).
System.String activity_regularizer

Regularizer function applied to the output of the layer (its 'activation'). (see regularizer).
System.String depthwise_constraint

Constraint function applied to the depthwise kernel matrix (see constraints).
System.String bias_constraint

Constraint function applied to the bias vector (see constraints).
Shape input_shape

4D tensor with shape: (batch, channels, rows, cols) if data_format is "channels_first" or 4D tensor with shape: (batch, rows, cols, channels) if data_format is "channels_last".

Implements

System.IDisposable