paul-corbalan
/
wasserstein-gan
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Merge pull request #69 from kopytjuk/bugfix-windows 

Bugfix to run the code on windows machines.
This commit is contained in:
Soumith Chintala 2018-12-25 11:43:31 -05:00 committed by GitHub
parent f81eafd2aa afd82ac354
commit 72853533a0
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 192 additions and 190 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

144
main.py
View File

@ -16,51 +16,53 @@ import os
import models.dcgan as dcgan import models.dcgan as dcgan
import models.mlp as mlp import models.mlp as mlp
parser = argparse.ArgumentParser() if __name__=="__main__":
parser.add_argument('--dataset', required=True, help='cifar10 | lsun | imagenet | folder | lfw ')
parser.add_argument('--dataroot', required=True, help='path to dataset')
parser.add_argument('--workers', type=int, help='number of data loading workers', default=2)
parser.add_argument('--batchSize', type=int, default=64, help='input batch size')
parser.add_argument('--imageSize', type=int, default=64, help='the height / width of the input image to network')
parser.add_argument('--nc', type=int, default=3, help='input image channels')
parser.add_argument('--nz', type=int, default=100, help='size of the latent z vector')
parser.add_argument('--ngf', type=int, default=64)
parser.add_argument('--ndf', type=int, default=64)
parser.add_argument('--niter', type=int, default=25, help='number of epochs to train for')
parser.add_argument('--lrD', type=float, default=0.00005, help='learning rate for Critic, default=0.00005')
parser.add_argument('--lrG', type=float, default=0.00005, help='learning rate for Generator, default=0.00005')
parser.add_argument('--beta1', type=float, default=0.5, help='beta1 for adam. default=0.5')
parser.add_argument('--cuda' , action='store_true', help='enables cuda')
parser.add_argument('--ngpu' , type=int, default=1, help='number of GPUs to use')
parser.add_argument('--netG', default='', help="path to netG (to continue training)")
parser.add_argument('--netD', default='', help="path to netD (to continue training)")
parser.add_argument('--clamp_lower', type=float, default=-0.01)
parser.add_argument('--clamp_upper', type=float, default=0.01)
parser.add_argument('--Diters', type=int, default=5, help='number of D iters per each G iter')
parser.add_argument('--noBN', action='store_true', help='use batchnorm or not (only for DCGAN)')
parser.add_argument('--mlp_G', action='store_true', help='use MLP for G')
parser.add_argument('--mlp_D', action='store_true', help='use MLP for D')
parser.add_argument('--n_extra_layers', type=int, default=0, help='Number of extra layers on gen and disc')
parser.add_argument('--experiment', default=None, help='Where to store samples and models')
parser.add_argument('--adam', action='store_true', help='Whether to use adam (default is rmsprop)')
opt = parser.parse_args()
print(opt)
if opt.experiment is None: parser = argparse.ArgumentParser()
parser.add_argument('--dataset', required=True, help='cifar10 | lsun | imagenet | folder | lfw ')
parser.add_argument('--dataroot', required=True, help='path to dataset')
parser.add_argument('--workers', type=int, help='number of data loading workers', default=2)
parser.add_argument('--batchSize', type=int, default=64, help='input batch size')
parser.add_argument('--imageSize', type=int, default=64, help='the height / width of the input image to network')
parser.add_argument('--nc', type=int, default=3, help='input image channels')
parser.add_argument('--nz', type=int, default=100, help='size of the latent z vector')
parser.add_argument('--ngf', type=int, default=64)
parser.add_argument('--ndf', type=int, default=64)
parser.add_argument('--niter', type=int, default=25, help='number of epochs to train for')
parser.add_argument('--lrD', type=float, default=0.00005, help='learning rate for Critic, default=0.00005')
parser.add_argument('--lrG', type=float, default=0.00005, help='learning rate for Generator, default=0.00005')
parser.add_argument('--beta1', type=float, default=0.5, help='beta1 for adam. default=0.5')
parser.add_argument('--cuda' , action='store_true', help='enables cuda')
parser.add_argument('--ngpu' , type=int, default=1, help='number of GPUs to use')
parser.add_argument('--netG', default='', help="path to netG (to continue training)")
parser.add_argument('--netD', default='', help="path to netD (to continue training)")
parser.add_argument('--clamp_lower', type=float, default=-0.01)
parser.add_argument('--clamp_upper', type=float, default=0.01)
parser.add_argument('--Diters', type=int, default=5, help='number of D iters per each G iter')
parser.add_argument('--noBN', action='store_true', help='use batchnorm or not (only for DCGAN)')
parser.add_argument('--mlp_G', action='store_true', help='use MLP for G')
parser.add_argument('--mlp_D', action='store_true', help='use MLP for D')
parser.add_argument('--n_extra_layers', type=int, default=0, help='Number of extra layers on gen and disc')
parser.add_argument('--experiment', default=None, help='Where to store samples and models')
parser.add_argument('--adam', action='store_true', help='Whether to use adam (default is rmsprop)')
opt = parser.parse_args()
print(opt)
if opt.experiment is None:
opt.experiment = 'samples' opt.experiment = 'samples'
os.system('mkdir {0}'.format(opt.experiment)) os.system('mkdir {0}'.format(opt.experiment))
opt.manualSeed = random.randint(1, 10000) # fix seed opt.manualSeed = random.randint(1, 10000) # fix seed
print("Random Seed: ", opt.manualSeed) print("Random Seed: ", opt.manualSeed)
random.seed(opt.manualSeed) random.seed(opt.manualSeed)
torch.manual_seed(opt.manualSeed) torch.manual_seed(opt.manualSeed)
cudnn.benchmark = True cudnn.benchmark = True
if torch.cuda.is_available() and not opt.cuda: if torch.cuda.is_available() and not opt.cuda:
print("WARNING: You have a CUDA device, so you should probably run with --cuda") print("WARNING: You have a CUDA device, so you should probably run with --cuda")
if opt.dataset in ['imagenet', 'folder', 'lfw']: if opt.dataset in ['imagenet', 'folder', 'lfw']:
# folder dataset # folder dataset
dataset = dset.ImageFolder(root=opt.dataroot, dataset = dset.ImageFolder(root=opt.dataroot,
transform=transforms.Compose([ transform=transforms.Compose([
@ -69,7 +71,7 @@ if opt.dataset in ['imagenet', 'folder', 'lfw']:
transforms.ToTensor(), transforms.ToTensor(),
transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5)), transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5)),
])) ]))
elif opt.dataset == 'lsun': elif opt.dataset == 'lsun':
dataset = dset.LSUN(db_path=opt.dataroot, classes=['bedroom_train'], dataset = dset.LSUN(db_path=opt.dataroot, classes=['bedroom_train'],
transform=transforms.Compose([ transform=transforms.Compose([
transforms.Scale(opt.imageSize), transforms.Scale(opt.imageSize),
@ -77,7 +79,7 @@ elif opt.dataset == 'lsun':
transforms.ToTensor(), transforms.ToTensor(),
transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5)), transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5)),
])) ]))
elif opt.dataset == 'cifar10': elif opt.dataset == 'cifar10':
dataset = dset.CIFAR10(root=opt.dataroot, download=True, dataset = dset.CIFAR10(root=opt.dataroot, download=True,
transform=transforms.Compose([ transform=transforms.Compose([
transforms.Scale(opt.imageSize), transforms.Scale(opt.imageSize),
@ -85,19 +87,19 @@ elif opt.dataset == 'cifar10':
transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5)), transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5)),
]) ])
) )
assert dataset assert dataset
dataloader = torch.utils.data.DataLoader(dataset, batch_size=opt.batchSize, dataloader = torch.utils.data.DataLoader(dataset, batch_size=opt.batchSize,
shuffle=True, num_workers=int(opt.workers)) shuffle=True, num_workers=int(opt.workers))
ngpu = int(opt.ngpu) ngpu = int(opt.ngpu)
nz = int(opt.nz) nz = int(opt.nz)
ngf = int(opt.ngf) ngf = int(opt.ngf)
ndf = int(opt.ndf) ndf = int(opt.ndf)
nc = int(opt.nc) nc = int(opt.nc)
n_extra_layers = int(opt.n_extra_layers) n_extra_layers = int(opt.n_extra_layers)
# custom weights initialization called on netG and netD # custom weights initialization called on netG and netD
def weights_init(m): def weights_init(m):
classname = m.__class__.__name__ classname = m.__class__.__name__
if classname.find('Conv') != -1: if classname.find('Conv') != -1:
m.weight.data.normal_(0.0, 0.02) m.weight.data.normal_(0.0, 0.02)
@ -105,51 +107,51 @@ def weights_init(m):
m.weight.data.normal_(1.0, 0.02) m.weight.data.normal_(1.0, 0.02)
m.bias.data.fill_(0) m.bias.data.fill_(0)
if opt.noBN: if opt.noBN:
netG = dcgan.DCGAN_G_nobn(opt.imageSize, nz, nc, ngf, ngpu, n_extra_layers) netG = dcgan.DCGAN_G_nobn(opt.imageSize, nz, nc, ngf, ngpu, n_extra_layers)
elif opt.mlp_G: elif opt.mlp_G:
netG = mlp.MLP_G(opt.imageSize, nz, nc, ngf, ngpu) netG = mlp.MLP_G(opt.imageSize, nz, nc, ngf, ngpu)
else: else:
netG = dcgan.DCGAN_G(opt.imageSize, nz, nc, ngf, ngpu, n_extra_layers) netG = dcgan.DCGAN_G(opt.imageSize, nz, nc, ngf, ngpu, n_extra_layers)
netG.apply(weights_init) netG.apply(weights_init)
if opt.netG != '': # load checkpoint if needed if opt.netG != '': # load checkpoint if needed
netG.load_state_dict(torch.load(opt.netG)) netG.load_state_dict(torch.load(opt.netG))
print(netG) print(netG)
if opt.mlp_D: if opt.mlp_D:
netD = mlp.MLP_D(opt.imageSize, nz, nc, ndf, ngpu) netD = mlp.MLP_D(opt.imageSize, nz, nc, ndf, ngpu)
else: else:
netD = dcgan.DCGAN_D(opt.imageSize, nz, nc, ndf, ngpu, n_extra_layers) netD = dcgan.DCGAN_D(opt.imageSize, nz, nc, ndf, ngpu, n_extra_layers)
netD.apply(weights_init) netD.apply(weights_init)
if opt.netD != '': if opt.netD != '':
netD.load_state_dict(torch.load(opt.netD)) netD.load_state_dict(torch.load(opt.netD))
print(netD) print(netD)
input = torch.FloatTensor(opt.batchSize, 3, opt.imageSize, opt.imageSize) input = torch.FloatTensor(opt.batchSize, 3, opt.imageSize, opt.imageSize)
noise = torch.FloatTensor(opt.batchSize, nz, 1, 1) noise = torch.FloatTensor(opt.batchSize, nz, 1, 1)
fixed_noise = torch.FloatTensor(opt.batchSize, nz, 1, 1).normal_(0, 1) fixed_noise = torch.FloatTensor(opt.batchSize, nz, 1, 1).normal_(0, 1)
one = torch.FloatTensor([1]) one = torch.FloatTensor([1])
mone = one * -1 mone = one * -1
if opt.cuda: if opt.cuda:
netD.cuda() netD.cuda()
netG.cuda() netG.cuda()
input = input.cuda() input = input.cuda()
one, mone = one.cuda(), mone.cuda() one, mone = one.cuda(), mone.cuda()
noise, fixed_noise = noise.cuda(), fixed_noise.cuda() noise, fixed_noise = noise.cuda(), fixed_noise.cuda()
# setup optimizer # setup optimizer
if opt.adam: if opt.adam:
optimizerD = optim.Adam(netD.parameters(), lr=opt.lrD, betas=(opt.beta1, 0.999)) optimizerD = optim.Adam(netD.parameters(), lr=opt.lrD, betas=(opt.beta1, 0.999))
optimizerG = optim.Adam(netG.parameters(), lr=opt.lrG, betas=(opt.beta1, 0.999)) optimizerG = optim.Adam(netG.parameters(), lr=opt.lrG, betas=(opt.beta1, 0.999))
else: else:
optimizerD = optim.RMSprop(netD.parameters(), lr = opt.lrD) optimizerD = optim.RMSprop(netD.parameters(), lr = opt.lrD)
optimizerG = optim.RMSprop(netG.parameters(), lr = opt.lrG) optimizerG = optim.RMSprop(netG.parameters(), lr = opt.lrG)
gen_iterations = 0 gen_iterations = 0
for epoch in range(opt.niter): for epoch in range(opt.niter):
data_iter = iter(dataloader) data_iter = iter(dataloader)
i = 0 i = 0
while i < len(dataloader): while i < len(dataloader):