mmagic.models.editors.stylegan3.stylegan3_modules

Module Contents

Classes

FullyConnectedLayer	Fully connected layer used in StyleGANv3.
MappingNetwork	Style mapping network used in StyleGAN3. The main difference between it
SynthesisInput	Module which generate input for synthesis layer.
SynthesisLayer	Layer of Synthesis network for stylegan3.
SynthesisNetwork	Synthesis network for stylegan3.

Functions

modulated_conv2d(x, w, s[, demodulate, padding, ...])

Modulated Conv2d in StyleGANv3.

Attributes

bias_act

mmagic.models.editors.stylegan3.stylegan3_modules.bias_act

mmagic.models.editors.stylegan3.stylegan3_modules.modulated_conv2d(x, w, s, demodulate=True, padding=0, input_gain=None)

Modulated Conv2d in StyleGANv3.

Parameters

• x (torch.Tensor) – Input tensor with shape (batch_size, in_channels, height, width).

• w (torch.Tensor) – Weight of modulated convolution with shape (out_channels, in_channels, kernel_height, kernel_width).

• s (torch.Tensor) – Style tensor with shape (batch_size, in_channels).

• demodulate (bool) – Whether apply weight demodulation. Defaults to True.

• padding (int or list[int]) – Convolution padding. Defaults to 0.

• input_gain (list[int]) – Scaling factors for input. Defaults to None.

Returns

Convolution Output.

Return type

torch.Tensor

class mmagic.models.editors.stylegan3.stylegan3_modules.FullyConnectedLayer(in_features, out_features, activation='linear', bias=True, lr_multiplier=1.0, weight_init=1.0, bias_init=0.0)

Bases: mmengine.model.BaseModule

Fully connected layer used in StyleGANv3.

Parameters

• in_features (int) – Number of channels in the input feature.

• out_features (int) – Number of channels in the out feature.

• activation (str, optional) – Activation function with choices ‘relu’, ‘lrelu’, ‘linear’. ‘linear’ means no extra activation. Defaults to ‘linear’.

• bias (bool, optional) – Whether to use additive bias. Defaults to True.

• lr_multiplier (float, optional) – Equalized learning rate multiplier. Defaults to 1..

• weight_init (float, optional) – Weight multiplier for initialization. Defaults to 1..

• bias_init (float, optional) – Initial bias. Defaults to 0..

forward(x)

Forward function.

class mmagic.models.editors.stylegan3.stylegan3_modules.MappingNetwork(noise_size, style_channels, num_ws, cond_size=0, num_layers=2, lr_multiplier=0.01, w_avg_beta=0.998)

Bases: mmengine.model.BaseModule

Style mapping network used in StyleGAN3. The main difference between it and styleganv1,v2 is that mean latent is registered as a buffer and dynamic updated during training.

Parameters

• noise_size (int, optional) – Size of the input noise vector.

• style_channels (int) – The number of channels for style code.

• num_ws (int | None) – The repeat times of w latent. If None is passed, the output will shape like (batch_size, 1), otherwise the output will shape like (bz, num_ws, 1).

• cond_size (int, optional) – Size of the conditional input. Defaults to None.

• num_layers (int, optional) – The number of layers of mapping network. Defaults to 2.

• lr_multiplier (float, optional) – Equalized learning rate multiplier. Defaults to 0.01.

• w_avg_beta (float, optional) – The value used for update w_avg. Defaults to 0.998.

forward(z, label=None, truncation=1, num_truncation_layer=None, update_emas=False)

Style mapping function.

Parameters

• z (torch.Tensor) – Input noise tensor.

• label (torch.Tensor, optional) – The conditional input. Defaults to None.

• truncation (float, optional) – Truncation factor. Give value less than 1., the truncation trick will be adopted. Defaults to 1.

• num_truncation_layer (int, optional) – Number of layers use truncated latent. Defaults to None.

• update_emas (bool, optional) – Whether update moving average of mean latent. Defaults to False.

Returns

W-plus latent.

Return type

torch.Tensor

class mmagic.models.editors.stylegan3.stylegan3_modules.SynthesisInput(style_channels, channels, size, sampling_rate, bandwidth)

Bases: mmengine.model.BaseModule

Module which generate input for synthesis layer.

Parameters

• style_channels (int) – The number of channels for style code.

• channels (int) – The number of output channel.

• size (int) – The size of sampling grid.

• sampling_rate (int) – Sampling rate for construct sampling grid.

• bandwidth (float) – Bandwidth of random frequencies.

forward(w)

Forward function.

class mmagic.models.editors.stylegan3.stylegan3_modules.SynthesisLayer(style_channels, is_torgb, is_critically_sampled, use_fp16, in_channels, out_channels, in_size, out_size, in_sampling_rate, out_sampling_rate, in_cutoff, out_cutoff, in_half_width, out_half_width, conv_kernel=3, filter_size=6, lrelu_upsampling=2, use_radial_filters=False, conv_clamp=256, magnitude_ema_beta=0.999)

Bases: mmengine.model.BaseModule

Layer of Synthesis network for stylegan3.

Parameters

• style_channels (int) – The number of channels for style code.

• is_torgb (bool) – Whether output of this layer is transformed to rgb image.

• is_critically_sampled (bool) – Whether filter cutoff is set exactly at the bandlimit.

• use_fp16 (bool, optional) – Whether to use fp16 training in this module. If this flag is True, the whole module will be wrapped with auto_fp16.

• in_channels (int) – The channel number of the input feature map.

• out_channels (int) – The channel number of the output feature map.

• in_size (int) – The input size of feature map.

• out_size (int) – The output size of feature map.

• in_sampling_rate (int) – Sampling rate for upsampling filter.

• out_sampling_rate (int) – Sampling rate for downsampling filter.

• in_cutoff (float) – Cutoff frequency for upsampling filter.

• out_cutoff (float) – Cutoff frequency for downsampling filter.

• in_half_width (float) – The approximate width of the transition region for upsampling filter.

• out_half_width (float) – The approximate width of the transition region for downsampling filter.

• conv_kernel (int, optional) – The kernel of modulated convolution. Defaults to 3.

• filter_size (int, optional) – Base filter size. Defaults to 6.

• lrelu_upsampling (int, optional) – Upsamling rate for filtered_lrelu. Defaults to 2.

• use_radial_filters (bool, optional) – Whether use radially symmetric jinc-based filter in downsamping filter. Defaults to False.

• conv_clamp (int, optional) – Clamp bound for convolution. Defaults to 256.

• magnitude_ema_beta (float, optional) – Beta coefficient for calculating input magnitude ema. Defaults to 0.999.

forward(x, w, force_fp32=False, update_emas=False)

Forward function for synthesis layer.

Parameters

• x (torch.Tensor) – Input feature map tensor.

• w (torch.Tensor) – Input style tensor.

• force_fp32 (bool, optional) – Force fp32 ignore the weights. Defaults to True.

• update_emas (bool, optional) – Whether update moving average of input magnitude. Defaults to False.

Returns

Output feature map tensor.

Return type

torch.Tensor

static design_lowpass_filter(numtaps, cutoff, width, fs, radial=False)

Design lowpass filter giving related arguments.

Parameters

• numtaps (int) – Length of the filter. numtaps must be odd if a passband includes the Nyquist frequency.

• cutoff (float) – Cutoff frequency of filter

• width (float) – The approximate width of the transition region.

• fs (float) – The sampling frequency of the signal.

• radial (bool, optional) – Whether use radially symmetric jinc-based filter. Defaults to False.

Returns

Kernel of lowpass filter.

Return type

torch.Tensor

class mmagic.models.editors.stylegan3.stylegan3_modules.SynthesisNetwork(style_channels, out_size, img_channels, channel_base=32768, channel_max=512, num_layers=14, num_critical=2, first_cutoff=2, first_stopband=2 ** 2.1, last_stopband_rel=2 ** 0.3, margin_size=10, output_scale=0.25, num_fp16_res=4, **layer_kwargs)

Bases: mmengine.model.BaseModule

Synthesis network for stylegan3.

Parameters

• style_channels (int) – The number of channels for style code.

• out_size (int) – The resolution of output image.

• img_channels (int) – The number of channels for output image.

• channel_base (int, optional) – Overall multiplier for the number of channels. Defaults to 32768.

• channel_max (int, optional) – Maximum number of channels in any layer. Defaults to 512.

• num_layers (int, optional) – Total number of layers, excluding Fourier features and ToRGB. Defaults to 14.

• num_critical (int, optional) – Number of critically sampled layers at the end. Defaults to 2.

• first_cutoff (int, optional) – Cutoff frequency of the first layer. Defaults to 2.

• first_stopband (int, optional) – Minimum stopband of the first layer. Defaults to 2**2.1.

• last_stopband_rel (float, optional) – Minimum stopband of the last layer, expressed relative to the cutoff. Defaults to 2**0.3.

• margin_size (int, optional) – Number of additional pixels outside the image. Defaults to 10.

• output_scale (float, optional) – Scale factor for output value. Defaults to 0.25.

• num_fp16_res (int, optional) – Number of first few layers use fp16. Defaults to 4.

forward(ws, **layer_kwargs)

Forward function.