transforms.Resize(opt.img_size),
transforms.ToTensor(),
transforms.Normalize([0.5], [0.5])
]),
),
batch_size=opt.batch_size,
shuffle=True,
)
os.makedirs("../../data/mnistm", exist_ok=True)
dataloader_B = torch.utils.data.DataLoader(
MNISTM(
"../../data/mnistm",
train=True,
download=True,
transform=transforms.Compose([
transforms.Resize(opt.img_size),
transforms.ToTensor(),
transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5)),
]),
),
batch_size=opt.batch_size,
shuffle=True,
)
# Optimizers
optimizer_G = torch.optim.Adam(itertools.chain(generator.parameters(),
classifier.parameters()),
lr=opt.lr,
betas=(opt.b1, opt.b2))
optimizer_D = torch.optim.Adam(discriminator.parameters(),
lr=opt.lr,
acc /= 10000.0
'''
return j, acc
net = FunctionJ()
if Use_GPU:
net = net.cuda(GPU_Num)
optimizer = opt.Adam(net.parameters(), lr=Learning_Rate, betas=(0.9, 0.99))
epoch = 0
transform = transforms.Compose([transforms.ToTensor()])
# dataset_train = datasets.MNIST(root='/media/data/zhaoyin/', transform=transform)
# dataset_test = datasets.MNIST(root='/media/data/zhaoyin/', train=False, transform=transform)
mnist_root = '/media/data/zhaoyin'
root = '/media/data/zhaoyin/mnistm'
dataset_train = MNISTM(root=root, mnist_root=mnist_root, transform=transform)
dataset_test = MNISTM(root=root,
mnist_root=mnist_root,
train=False,
transform=transform)
loss_ls = []
acc_ls = []
def isfinish(li):
fi = 0
for t in range(len(li) - 1):
fi += abs(li[t + 1] - li[t])
if fi < li[-1] / 20:
return True
train=True,
download=True,
transform=transforms.Compose([
transforms.Resize(opt.img_size),
transforms.ToTensor(),
transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))
])),
batch_size=opt.batch_size,
shuffle=True)
os.makedirs('../../data/mnistm', exist_ok=True)
dataloader_B = torch.utils.data.DataLoader(MNISTM(
'../../data/mnistm',
train=True,
download=True,
transform=transforms.Compose([
transforms.Resize(opt.img_size),
transforms.ToTensor(),
transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))
])),
batch_size=opt.batch_size,
shuffle=True)
# Optimizers
optimizer_G = torch.optim.Adam(itertools.chain(generator.parameters(),
classifier.parameters()),
lr=opt.lr,
betas=(opt.b1, opt.b2))
optimizer_D = torch.optim.Adam(discriminator.parameters(),
lr=opt.lr,
betas=(opt.b1, opt.b2))
dataset = torchvision.datasets.STL10(root=dset_dir,
download=True,
split='train')
print(dataset.data.shape)
print(dataset.data.mean(axis=(0, 2, 3)) / 255)
print(dataset.data.std(axis=(0, 2, 3)) / 255)
elif dset_name == "mnist":
dataset = torchvision.datasets.MNIST(root=dset_dir, train=True)
print(list(dataset.train_data.size()))
print(dataset.train_data.float().mean() / 255)
print(dataset.train_data.float().std() / 255)
elif dset_name == "mnistm":
from mnistm import MNISTM
dataset = MNISTM(root=dset_dir, train=True)
print(list(dataset.train_data.size()))
for dim in range(3):
print(dim)
print(dataset.train_data[:, :, :, dim].float().mean() / 255)
print(dataset.train_data[:, :, :, dim].float().std() / 255)
elif dset_name == "svhn":
dataset = torchvision.datasets.SVHN(root=dset_dir,
download=True,
split='train')
print(dataset.data.shape)
print(dataset.data.mean(axis=(0, 2, 3)) / 255)
print(dataset.data.std(axis=(0, 2, 3)) / 255)
elif dset_name == "usps":