mmagic.models.editors.wgan_gp.wgan_discriminator

Module Contents

Classes

WGANGPDiscriminator

Discriminator for WGANGP.

class mmagic.models.editors.wgan_gp.wgan_discriminator.WGANGPDiscriminator(in_channel, in_scale, conv_module_cfg=None, init_cfg=None)

Bases: mmengine.model.BaseModule

Discriminator for WGANGP.

Implementation Details for WGANGP discriminator the same as training configuration (a) described in PGGAN paper: PROGRESSIVE GROWING OF GANS FOR IMPROVED QUALITY, STABILITY, AND VARIATION https://research.nvidia.com/sites/default/files/pubs/2017-10_Progressive-Growing-of/karras2018iclr-paper.pdf # noqa

1. Adopt convolution architecture specified in appendix A.2;

2. Add layer normalization to all conv3x3 and conv4x4 layers;

3. Use LeakyReLU in the discriminator except for the final output layer;

4. Initialize all weights using He’s initializer.

Parameters

• in_channel (int) – The channel number of the input image.

• in_scale (int) – The scale of the input image.

• conv_module_cfg (dict, optional) – Config for the convolution module used in this discriminator. Defaults to None.

• init_cfg (dict, optional) – Initialization config dict.

_default_channels_per_scale

_default_conv_module_cfg

_default_upsample_cfg

forward(x)

Forward function.

Parameters

x (torch.Tensor) – Fake or real image tensor.

Returns

Prediction for the reality of the input image.

Return type

torch.Tensor