dataset = NewDogCat(root=opt.dataRoot, transform=transform)
dataloader = torch.utils.data.DataLoader(dataset,
batch_size=opt.batchSize,
collate_fn=my_collate_fn,
shuffle=True,
num_workers=opt.workers)
# load models
if opt.model == 1:
netG = models._netG_1(ngpu, nz, nc, ngf, n_extra_g)
netD = models._netD_1(ngpu, nz, nc, ndf, n_extra_d)
elif opt.model == 2:
netG = models._netG_2(ngpu, nz, nc, ngf)
netD = models._netD_2(ngpu, nz, nc, ndf)
netG.apply(weights_init)
if opt.netG != '':
netG.load_state_dict(torch.load(opt.netG))
print(netG)
netD.apply(weights_init)
if opt.netD != '':
netD.load_state_dict(torch.load(opt.netD))
print(netD)
criterion = nn.BCELoss()
criterion_MSE = nn.MSELoss()
input = torch.FloatTensor(opt.batchSize, 3, opt.imageSize, opt.imageSize)
transforms.Scale(opt.imageSize),
# transforms.CenterCrop(opt.imageSize),
transforms.ToTensor(),
transforms.Normalize((0.5,0.5,0.5), (0.5,0.5,0.5)), # bring images to (-1,1)
])
)
dataloader = torch.utils.data.DataLoader(dataset, batch_size=opt.batchSize,
shuffle=True, num_workers=opt.workers)
# load models
if opt.model == 1:
netG = models._netG_1(ngpu, nz, nc, ngf, n_extra_g)
netD = models._netD_1(ngpu, nz, nc, ndf, n_extra_d)
elif opt.model == 2:
netG = models._netG_2(ngpu, nz, nc, ngf)
netD = models._netD_2(ngpu, nz, nc, ndf)
netG.apply(weights_init)
if opt.netG != '':
netG.load_state_dict(torch.load(opt.netG))
print(netG)
netD.apply(weights_init)
if opt.netD != '':
netD.load_state_dict(torch.load(opt.netD))
print(netD)
criterion = nn.BCELoss()
criterion_MSE = nn.MSELoss()
input = torch.FloatTensor(opt.batchSize, 3, opt.imageSize, opt.imageSize)