batch_size = args.batch_size
n_classes = 10
input_channels = 1
seq_length = int(784 / input_channels)
epochs = args.epochs
steps = 0
print(args)
train_loader, test_loader = data_generator(root, batch_size)
permute = torch.Tensor(np.random.permutation(784).astype(np.float64)).long()
channel_sizes = [args.nhid] * args.levels
kernel_size = args.ksize
model = TCN(input_channels,
n_classes,
channel_sizes,
kernel_size=kernel_size,
dropout=args.dropout)
if args.cuda:
model.cuda()
permute = permute.cuda()
lr = args.lr
optimizer = getattr(optim, args.optim)(model.parameters(), lr=lr)
def train(ep):
global steps
train_loss = 0
model.train()
iterations = 0
test_acc = []
print(args)
train_loader, test_loader = data_generator(root, batch_size)
permute = torch.Tensor(np.random.permutation(784).astype(np.float64)).long()
channel_sizes = [args.nhid] * args.levels
kernel_size = args.ksize
if args.lstm:
model = LSTM(input_channels, 75, n_classes)
else:
model = TCN(input_channels,
n_classes,
channel_sizes,
kernel_size=kernel_size,
dropout=args.dropout)
print(count_parameters(model))
if args.cuda:
model.cuda()
permute = permute.cuda()
lr = args.lr
optimizer = getattr(optim, args.optim)(model.parameters(), lr=lr)
def train(ep):
global steps, iterations, test_acc
root = './data/mnist'
batch_size = args.batch_size
n_classes = 10
input_channels = 1
seq_length = int(784 / input_channels)
epochs = args.epochs
steps = 0
print(args)
train_loader, test_loader = data_generator(root, batch_size)
permute = torch.Tensor(np.random.permutation(784).astype(np.float64)).long()
channel_sizes = [args.nhid] * args.levels
kernel_size = args.ksize
model = TCN(input_channels, n_classes, channel_sizes, kernel_size=kernel_size, dropout=args.dropout)
if args.cuda:
model.cuda()
permute = permute.cuda()
lr = args.lr
optimizer = getattr(optim, args.optim)(model.parameters(), lr=lr)
def train(ep):
global steps
train_loss = 0
model.train()
for batch_idx, (data, target) in enumerate(train_loader):
if args.cuda: data, target = data.cuda(), target.cuda()
batch_size = args.batch_size
n_classes = 10
input_channels = 1
seq_length = int(784 / input_channels)
epochs = args.epochs
steps = 0
print(args)
train_loader, test_loader = data_generator(root, batch_size)
permute = torch.Tensor(np.random.permutation(784).astype(np.float64)).long()
channel_sizes = [args.nhid] * args.levels
kernel_size = args.ksize
model = TCN(input_channels,
n_classes,
channel_sizes,
kernel_size=kernel_size,
dropout=args.dropout)
if args.cuda:
model.cuda()
permute = permute.cuda()
lr = args.lr
optimizer = getattr(optim, args.optim)(model.parameters(), lr=lr)
def train(ep):
global steps
train_loss = 0
epoch_loss = []
batch_size = args.batch_size
n_classes = 10
input_channels = 1
seq_length = int(784 / input_channels)
epochs = args.epochs
steps = 0
print(args)
train_loader, test_loader = data_generator(root, batch_size)
permute = torch.Tensor(np.random.permutation(784).astype(np.float64)).long()
channel_sizes = [args.nhid] * args.levels
kernel_size = args.ksize
model = TCN(input_channels,
n_classes,
channel_sizes,
kernel_size=kernel_size,
dropout=args.dropout)
model = Chrysalis.metamorphosize(model, in_place=True)
if args.patch_conv:
X = next(iter(train_loader))
model.patch_conv(
X[0].view(-1, input_channels, seq_length)[:1],
verbose=True,
kmatrix_depth=args.kmatrix_depth,
max_kernel_size=args.max_kernel_size,
padding_mode='zeros',
base=args.base,
perturb=args.perturb,
)
help='directory to save model')
parser.add_argument('--batch_size',
type=int,
default=64,
metavar='N',
help='batch size (default: 64)')
args = parser.parse_args()
if torch.cuda.is_available():
if not args.cuda:
print(
"WARNING: You have a CUDA device, so you should probably run with --cuda"
)
_, test_loader = data_generator(args.datapath, args.batch_size)
model = TCN()
model.load_state_dict(torch.load(args.modelpath))
model.eval()
if args.cuda:
model.cuda()
model.fast_inference(args.batch_size)
def test():
test_loss = 0
correct = 0
with torch.no_grad():
for data, target in test_loader:
if args.cuda: