Welcome to our Support Center

Convolution 3D Transpose

Description

Define the convolution 3D transpose layer according to its parameters. To be used for the TimeDistributed layer. Type : polymorphic.

 

Input parameters

 

parameters : layer parameters.

 n_filters : integer, the dimensionality of the output space.
Default value “3”.
 size : array integer, specify the depth, height and width of the 3D convolution window. Can be a single integer to specify the same value for all spatial dimensions.
Default value “[3,3,3]”. Never more 3 values
 stride : array integer, specify the strides of the convolution along the depth, height and width. Can be a single integer to specify the same value for all spatial dimensions.
Default value “[1,1,1]”. Never more 3 values
dilation_rate : integer, specifying the dilation rate to use for dilated convolution.
Default value “[1,1,1]”. Never more 3 values
 activation : enum, activation function to use.
Default value “relu”.
 use_bias? : boolean, whether the layer uses a bias vector.
Default value “True”.
 padding : boolean, False = “valid” means no padding. True = “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.
Default value “False”.
 data_format : enum, one of channels_last or channels_first (default) . The ordering of the dimensions in the inputs. channel_last corresponds to inputs with shape (batch, steps, features) while channels_first corresponds to inputs with shape (batch, features, steps).
Default value “channels_first”.
 optimizer :

 algorithm : enum, (name of optimizer) for optimizer instance.
Default value “adam”.
 learning_rate : float, define the learning rate to use.
Default value “0.001”.
 beta_1 : float, define the exponential decay rate for the 1st moment estimates.
Default value “0.9”.
 beta_2 : float, define the exponential decay rate for the 2nd moment estimates.
Default value “0.999”.

 filter_initializer : enum, initializer for the kernel weights matrix.
Default value “glorot_uniform”.
 bias_initializer : enum, initializer for the bias vector.
Default value “zero”.
 training? : boolean, whether the layer is in training mode (can store data for backward).
Default value “True”.
 store? : boolean, whether the layer stores the last iteration gradient (accessible via the “get_gradients” function).
Default value “False”.
 update? : boolean, whether the layer’s variables should be updated during backward. Equivalent to freeze the layer.
Default value “True”.
 lda_coeff : float, defines the coefficient by which the loss derivative will be multiplied before being sent to the previous layer (since during the backward run we go backwards).
Default value “1”.

 

Output parameters

 

Conv3DTranspose out : layer convolution 3D transpose architecture.

Example

All these exemples are snippets PNG, you can drop these Snippet onto the block diagram and get the depicted code added to your VI (Do not forget to install HAIBAL library to run it).

Convolution 3D Transpose layer inside TimeDistributed layer

1 – Generate a set of data

We generate an array of data of type single and shape [batch_size = 10, time = 6, channel = 5, conv_dim1 = 64, conv_dim2 = 64, conv_dim3 = 3] (channel first default layer configuration).
In case of channel last layer configuration, shape is [batch_size, time, conv_dim1, conv_dim2, conv_dim3, channel].

2 – Define graph

First, we define the first layer of the graph which is an Input layer (explicit input layer method). This layer is setup as an input array shaped [time = 6, channel = 5, conv_dim1 = 64, conv_dim2 = 64, conv_dim3 = 3].
Then, we add to the graph the TimeDistributed layer which we setup with a Conv3DTranspose layer using the define method.

3 – Run graph

We call the forward method and retrieve the result with the “Prediction 6D” method.
This method returns two variables, the first one is the layer information (cluster composed of the layer name, the graph index and the shape of the output layer) and the second one is the prediction with a shape of [batch_size, time, new_conv_dim1, new_conv_dim2, new_conv_dim3].

 

Table of Contents