Spaces:
Running
Running
| import functools | |
| import torch.nn as nn | |
| from ..basic import ActNorm, CircularConv2d | |
| class NLayerDiscriminator(nn.Module): | |
| """Defines a PatchGAN discriminator as in Pix2Pix | |
| --> see https://github.com/junyanz/pytorch-CycleGAN-and-pix2pix/blob/master/models/networks.py | |
| """ | |
| def __init__(self, input_nc=1, output_nc=1, ndf=64, n_layers=3, use_actnorm=False): | |
| """Construct a PatchGAN discriminator | |
| Parameters: | |
| input_nc (int) -- the number of channels in input images | |
| ndf (int) -- the number of filters in the last conv layer | |
| n_layers (int) -- the number of conv layers in the discriminator | |
| norm_layer -- normalization layer | |
| """ | |
| super(NLayerDiscriminator, self).__init__() | |
| if not use_actnorm: | |
| norm_layer = nn.BatchNorm2d | |
| else: | |
| norm_layer = ActNorm | |
| if type(norm_layer) == functools.partial: # no need to use bias as BatchNorm2d has affine parameters | |
| use_bias = norm_layer.func != nn.BatchNorm2d | |
| else: | |
| use_bias = norm_layer != nn.BatchNorm2d | |
| kw = 4 | |
| padw = 1 | |
| sequence = [nn.Conv2d(input_nc, ndf, kernel_size=kw, stride=2, padding=padw), nn.LeakyReLU(0.2, True)] | |
| nf_mult = 1 | |
| for n in range(1, n_layers): # gradually increase the number of filters | |
| nf_mult_prev = nf_mult | |
| nf_mult = min(2 ** n, 8) | |
| sequence += [ | |
| nn.Conv2d(ndf * nf_mult_prev, ndf * nf_mult, kernel_size=kw, stride=2, padding=padw, bias=use_bias), | |
| norm_layer(ndf * nf_mult), | |
| nn.LeakyReLU(0.2, True) | |
| ] | |
| nf_mult_prev = nf_mult | |
| nf_mult = min(2 ** n_layers, 8) | |
| sequence += [ | |
| nn.Conv2d(ndf * nf_mult_prev, ndf * nf_mult, kernel_size=kw, stride=1, padding=padw, bias=use_bias), | |
| norm_layer(ndf * nf_mult), | |
| nn.LeakyReLU(0.2, True) | |
| ] | |
| sequence += [ | |
| nn.Conv2d(ndf * nf_mult, output_nc, kernel_size=kw, stride=1, padding=padw)] # output 1 channel prediction map | |
| self.main = nn.Sequential(*sequence) | |
| def forward(self, input): | |
| """Standard forward.""" | |
| return self.main(input) | |
| class LiDARNLayerDiscriminator(nn.Module): | |
| """Modified PatchGAN discriminator from Pix2Pix | |
| --> see https://github.com/junyanz/pytorch-CycleGAN-and-pix2pix/blob/master/models/networks.py | |
| """ | |
| def __init__(self, input_nc=1, output_nc=1, ndf=64, n_layers=3, use_actnorm=False): | |
| """Construct a PatchGAN discriminator | |
| Parameters: | |
| input_nc (int) -- the number of channels in input images | |
| ndf (int) -- the number of filters in the last conv layer | |
| n_layers (int) -- the number of conv layers in the discriminator | |
| norm_layer -- normalization layer | |
| """ | |
| super(LiDARNLayerDiscriminator, self).__init__() | |
| if not use_actnorm: | |
| norm_layer = nn.BatchNorm2d | |
| else: | |
| norm_layer = ActNorm | |
| if type(norm_layer) == functools.partial: # no need to use bias as BatchNorm2d has affine parameters | |
| use_bias = norm_layer.func != nn.BatchNorm2d | |
| else: | |
| use_bias = norm_layer != nn.BatchNorm2d | |
| kw = (4, 4) | |
| sequence = [CircularConv2d(input_nc, ndf, kernel_size=kw, stride=(1, 2), padding=(1, 2, 1, 2)), nn.LeakyReLU(0.2, True)] | |
| nf_mult = 1 | |
| nf_mult_prev = 1 | |
| for n in range(1, n_layers): # gradually increase the number of filters | |
| nf_mult_prev = nf_mult | |
| nf_mult = min(2 ** n, 8) | |
| sequence += [ | |
| CircularConv2d(ndf * nf_mult_prev, ndf * nf_mult, kernel_size=kw, stride=(1, 2), bias=use_bias, padding=(1, 2, 1, 2)), | |
| norm_layer(ndf * nf_mult), | |
| nn.LeakyReLU(0.2, True) | |
| ] | |
| nf_mult_prev = nf_mult | |
| nf_mult = min(2 ** n_layers, 8) | |
| sequence += [ | |
| CircularConv2d(ndf * nf_mult_prev, ndf * nf_mult, kernel_size=kw, stride=1, bias=use_bias, padding=(1, 2, 1, 2)), | |
| norm_layer(ndf * nf_mult), | |
| nn.LeakyReLU(0.2, True) | |
| ] | |
| sequence += [ | |
| CircularConv2d(ndf * nf_mult, output_nc, kernel_size=kw, stride=1, padding=(1, 2, 1, 2))] # output 1 channel prediction map | |
| self.main = nn.Sequential(*sequence) | |
| def forward(self, input): | |
| """Standard forward.""" | |
| return self.main(input) | |
| class LiDARNLayerDiscriminatorV2(nn.Module): | |
| """Modified PatchGAN discriminator from Pix2Pix (larger receptive field) | |
| --> see https://github.com/junyanz/pytorch-CycleGAN-and-pix2pix/blob/master/models/networks.py | |
| """ | |
| def __init__(self, input_nc=1, output_nc=1, ndf=64, n_layers=3, use_actnorm=False): | |
| """Construct a PatchGAN discriminator | |
| Parameters: | |
| input_nc (int) -- the number of channels in input images | |
| ndf (int) -- the number of filters in the last conv layer | |
| n_layers (int) -- the number of conv layers in the discriminator | |
| norm_layer -- normalization layer | |
| """ | |
| super(LiDARNLayerDiscriminatorV2, self).__init__() | |
| if not use_actnorm: | |
| norm_layer = nn.BatchNorm2d | |
| else: | |
| norm_layer = ActNorm | |
| if type(norm_layer) == functools.partial: # no need to use bias as BatchNorm2d has affine parameters | |
| use_bias = norm_layer.func != nn.BatchNorm2d | |
| else: | |
| use_bias = norm_layer != nn.BatchNorm2d | |
| kw = (4, 4) | |
| sequence = [CircularConv2d(input_nc, ndf, kernel_size=kw, stride=(1, 2), padding=(1, 2, 1, 2)), nn.LeakyReLU(0.2, True), | |
| CircularConv2d(ndf, ndf, kernel_size=kw, stride=(1, 2), padding=(1, 2, 1, 2)), nn.LeakyReLU(0.2, True)] | |
| nf_mult = 1 | |
| nf_mult_prev = 1 | |
| for n in range(1, n_layers): # gradually increase the number of filters | |
| nf_mult_prev = nf_mult | |
| nf_mult = min(2 ** n, 8) | |
| sequence += [ | |
| CircularConv2d(ndf * nf_mult_prev, ndf * nf_mult, kernel_size=kw, stride=(2, 2), bias=use_bias, padding=(1, 2, 1, 2)), | |
| norm_layer(ndf * nf_mult), | |
| nn.LeakyReLU(0.2, True) | |
| ] | |
| nf_mult_prev = nf_mult | |
| nf_mult = min(2 ** n_layers, 8) | |
| sequence += [ | |
| CircularConv2d(ndf * nf_mult_prev, ndf * nf_mult, kernel_size=kw, stride=1, bias=use_bias, padding=(1, 2, 1, 2)), | |
| norm_layer(ndf * nf_mult), | |
| nn.LeakyReLU(0.2, True) | |
| ] | |
| sequence += [ | |
| CircularConv2d(ndf * nf_mult, output_nc, kernel_size=kw, stride=1, padding=(1, 2, 1, 2))] # output 1 channel prediction map | |
| self.main = nn.Sequential(*sequence) | |
| def forward(self, input): | |
| """Standard forward.""" | |
| return self.main(input) | |
| class LiDARNLayerDiscriminatorV3(nn.Module): | |
| """Modified PatchGAN discriminator from Pix2Pix (larger receptive field) | |
| --> see https://github.com/junyanz/pytorch-CycleGAN-and-pix2pix/blob/master/models/networks.py | |
| """ | |
| def __init__(self, input_nc=1, output_nc=1, ndf=64, n_layers=3, use_actnorm=False): | |
| """Construct a PatchGAN discriminator | |
| Parameters: | |
| input_nc (int) -- the number of channels in input images | |
| ndf (int) -- the number of filters in the last conv layer | |
| n_layers (int) -- the number of conv layers in the discriminator | |
| norm_layer -- normalization layer | |
| """ | |
| super(LiDARNLayerDiscriminatorV3, self).__init__() | |
| if not use_actnorm: | |
| norm_layer = nn.BatchNorm2d | |
| else: | |
| norm_layer = ActNorm | |
| if type(norm_layer) == functools.partial: # no need to use bias as BatchNorm2d has affine parameters | |
| use_bias = norm_layer.func != nn.BatchNorm2d | |
| else: | |
| use_bias = norm_layer != nn.BatchNorm2d | |
| kw = (4, 4) | |
| sequence = [CircularConv2d(input_nc, ndf, kernel_size=(1, 4), stride=(1, 1), padding=(1, 2, 1, 2)), nn.LeakyReLU(0.2, True), | |
| CircularConv2d(ndf, ndf, kernel_size=kw, stride=(2, 2), padding=(1, 2, 1, 2)), nn.LeakyReLU(0.2, True)] | |
| nf_mult = 1 | |
| nf_mult_prev = 1 | |
| for n in range(1, n_layers): # gradually increase the number of filters | |
| nf_mult_prev = nf_mult | |
| nf_mult = min(2 ** n, 8) | |
| sequence += [ | |
| CircularConv2d(ndf * nf_mult_prev, ndf * nf_mult, kernel_size=kw, stride=(2, 2), bias=use_bias, padding=(1, 2, 1, 2)), | |
| norm_layer(ndf * nf_mult), | |
| nn.LeakyReLU(0.2, True) | |
| ] | |
| nf_mult_prev = nf_mult | |
| nf_mult = min(2 ** n_layers, 8) | |
| sequence += [ | |
| CircularConv2d(ndf * nf_mult_prev, ndf * nf_mult, kernel_size=kw, stride=1, bias=use_bias, padding=(1, 2, 1, 2)), | |
| norm_layer(ndf * nf_mult), | |
| nn.LeakyReLU(0.2, True) | |
| ] | |
| sequence += [ | |
| CircularConv2d(ndf * nf_mult, output_nc, kernel_size=kw, stride=1, padding=(1, 2, 1, 2))] # output 1 channel prediction map | |
| self.main = nn.Sequential(*sequence) | |
| def forward(self, input): | |
| """Standard forward.""" | |
| import pdb; pdb.set_trace() | |
| return self.main(input) | |