pysegcnn.core.layers

Layers of a convolutional encoder-decoder network.

License

Copyright (c) 2020 Daniel Frisinghelli

This source code is licensed under the GNU General Public License v3.

See the LICENSE file in the repository’s root directory.

Functions

conv_bn_relu(in_channels, out_channels, **kwargs)

Block of convolution, batch normalization and rectified linear unit.

Classes

Block(in_channels, out_channels, **kwargs)

Basic convolutional block.

Conv2dSame(*args, **kwargs)

A convolution preserving the shape of its input.

ConvBnReluMaxPool(in_channels, out_channels, …)

Block of convolution, batchnorm, relu and 2x2 max pool.

ConvBnReluMaxUnpool(in_channels, …)

Block of convolution, batchnorm, relu and 2x2 max unpool.

ConvBnReluUpsample(in_channels, …)

Block of convolution, batchnorm, relu and nearest neighbor upsampling.

Decoder(filters, block[, skip])

Generic convolutional decoder.

DecoderBlock(in_channels, out_channels, **kwargs)

Block of a convolutional decoder.

Encoder(filters, block, **kwargs)

Generic convolutional encoder.

EncoderBlock(in_channels, out_channels, **kwargs)

Block of a convolutional encoder.