0 ×

Keras Transposed Convolution 2D Layer

KNIME Deep Learning - Keras Integration version 3.6.0.v201807091039 by KNIME AG, Zurich, Switzerland

The need for transposed convolutions generally arises from the desire to use a transformation going in the opposite direction of a normal convolution, i.e., from something that has the shape of the output of some convolution to something that has the shape of its input while maintaining a connectivity pattern that is compatible with said convolution. Corresponds to the Keras Transposed Convolution 2D Layer.

Options

Name prefix
The name prefix of the layer. The prefix is complemented by an index suffix to obtain a unique layer name. If this option is unchecked, the name prefix is derived from the layer type.
Filters
The dimensionality of the output space (i.e. the number of output filters in the convolution).
Kernel size
A tuple of 2 integers, specifying the height and width of the 2D convolution window.
Strides
A tuple of 2 integers, specifying the strides of the convolution along the height and width. Specifying any stride value != 1 is incompatible with specifying any dilation_rate value != 1.
Padding
Different padding modes to apply to the spatial dimensions (excluding the batch and channel dimensions) of the inputs before the pooling operation. The padding will be done with zeroes. A detailed explanation of the different modes can be found here .
  • Valid: No padding
  • Same: Padding such that the spatial output dimension do not change.
  • Full: Padding with kernel size - 1
Dilation rate
A tuple/list of 2 integers, 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 function
The activation function to use.
Use bias?
If checked, a bias vector will be used.
Kernel initializer
Initializer for the kernel weights matrix.
Bias initializer
Initializer for the bias vector.
Kernel regularizer
Regularizer function applied to the kernel weights matrix.
Bias regularizer
Regularizer function applied to the bias vector.
Activation regularizer
Regularizer function applied to the output of the layer (its "activation").
Kernel constraint
Constraint function applied to the kernel matrix.
Bias constraint
Constraint function applied to the bias vector.

Input Ports

The Keras deep learning network to which to add a Transposed Convolution 2D layer.

Output Ports

The Keras deep learning network with an added Transposed Convolution 2D layer.

Update Site

To use this node in KNIME, install KNIME Deep Learning - Keras Integration from the following update site:

Wait a sec! You want to explore and install nodes even faster? We highly recommend our NodePit for KNIME extension for your KNIME Analytics Platform.