def train(**kwargs):
conf.parse(kwargs)
# train_set = DataSet(cfg, train=True, test=False)
train_set = ImageFolder(conf.TRAIN_DATA_ROOT, transform)
train_loader = DataLoader(train_set, conf.BATCH_SIZE,
shuffle=True,
num_workers=conf.NUM_WORKERS)
model = Network()
if conf.LOAD_MODEL_PATH:
print(conf.LOAD_MODEL_PATH)
model.load_state_dict(torch.load(conf.CHECKPOINTS_ROOT + conf.LOAD_MODEL_PATH))
device = torch.device('cuda:0' if conf.USE_GPU else 'cpu')
criterion = nn.CrossEntropyLoss().to(device)
lr = conf.LEARNING_RATE
optim = torch.optim.Adam(params=model.parameters(),
lr=lr,
weight_decay=conf.WEIGHT_DECAY)
model.to(device)
for epoch in range(conf.MAX_EPOCH):
model.train()
running_loss = 0
for step, (inputs, targets) in tqdm(enumerate(train_loader)):
inputs, targets = inputs.to(device), targets.to(device)
optim.zero_grad()
outs = model(inputs)
loss = criterion(outs, targets)
loss.backward()
optim.step()
running_loss += loss.item()
if step % conf.PRINT_FREQ == conf.PRINT_FREQ - 1:
running_loss = running_loss / conf.PRINT_FREQ
print('[%d, %5d] loss: %.3f' % (epoch + 1, step + 1, running_loss))
# vis.plot('loss', running_loss)
running_loss = 0
torch.save(model.state_dict(), conf.CHECKPOINTS_ROOT + time.strftime('%Y-%m-%d-%H-%M-%S.pth'))
for param_group in optim.param_groups:
lr *= conf.LEARNING_RATE_DECAY
param_group['lr'] = lr
###############################################################################
print(f"{gct()} : Start training")
best_ms = None
best_f = None
start_epoch = args.start_epoch + 1
end = cfg.TRAIN.EPOCH_NUM
for epoch in range(start_epoch, end):
epoch_start_time = time.time()
train()
checkpoint, val_ms = evaluate(val_data)
# Save the model if the match score is the best we've seen so far.
if not best_ms or val_ms >= best_ms:
state = {
"epoch": epoch,
"state_dict": model.state_dict(),
"det_optim": det_optim.state_dict(),
"des_optim": des_optim.state_dict(),
}
filename = f"{args.save}/model/e{epoch:03d}_{checkpoint}.pth.tar"
torch.save(state, filename)
best_ms = val_ms
best_f = filename
print("-" * 96)
print(
"| end of epoch {:3d} | time: {:5.02f}s | val ms {:5.03f} | best ms {:5.03f} | "
.format(epoch, (time.time() - epoch_start_time), val_ms, best_ms))
print("-" * 96)
# Load the best saved model.
def main():
train_dataset = MNIST(root='./data',
train=True,
download=True,
transform=transforms.ToTensor())
test_dataset = MNIST(root='./data',
train=False,
download=True,
transform=transforms.ToTensor())
train_loader = DataLoader(train_dataset,
batch_size=BATCH_SIZE,
shuffle=True,
num_workers=2)
test_loader = DataLoader(test_dataset,
batch_size=BATCH_SIZE,
shuffle=False,
num_workers=2)
net = Network(1, 64, 5, 10)
if USE_CUDA:
net = net.cuda()
opt = optim.SGD(net.parameters(),
lr=LEARNING_RATE,
weight_decay=WEIGHT_DECAY,
momentum=.9,
nesterov=True)
if not os.path.exists('checkpoint'):
os.mkdir('checkpoint')
for epoch in range(1, EPOCHS + 1):
print('[Epoch %d]' % epoch)
train_loss = 0
train_correct, train_total = 0, 0
start_point = time.time()
for inputs, labels in train_loader:
inputs, labels = Variable(inputs), Variable(labels)
if USE_CUDA:
inputs, labels = inputs.cuda(), labels.cuda()
opt.zero_grad()
preds = F.log_softmax(net(inputs), dim=1)
loss = F.cross_entropy(preds, labels)
loss.backward()
opt.step()
train_loss += loss.item()
train_correct += (preds.argmax(dim=1) == labels).sum().item()
train_total += len(preds)
print('train-acc : %.4f%% train-loss : %.5f' %
(100 * train_correct / train_total,
train_loss / len(train_loader)))
print('elapsed time: %ds' % (time.time() - start_point))
test_loss = 0
test_correct, test_total = 0, 0
for inputs, labels in test_loader:
with torch.no_grad():
inputs, labels = Variable(inputs), Variable(labels)
if USE_CUDA:
inputs, labels = inputs.cuda(), labels.cuda()
preds = F.softmax(net(inputs), dim=1)
test_loss += F.cross_entropy(preds, labels).item()
test_correct += (preds.argmax(dim=1) == labels).sum().item()
test_total += len(preds)
print('test-acc : %.4f%% test-loss : %.5f' %
(100 * test_correct / test_total, test_loss / len(test_loader)))
torch.save(net.state_dict(),
'./checkpoint/checkpoint-%04d.bin' % epoch)