def train_model(modname='alexnet', pm_ch='both', bs=16):
"""
Args:
modname (string): Name of the model. Has to be one of the values:
'alexnet', batch 64
'densenet'
'inception'
'resnet', batch 16
'squeezenet', batch 16
'vgg'
pm_ch (string): pixelmap channel -- 'time', 'charge', 'both', default to both
"""
# device configuration
device = torch.device('cuda:0' if torch.cuda.is_available() else 'cpu')
# hyper parameters
max_epochs = 10
learning_rate = 0.001
# determine number of input channels
nch = 2
if pm_ch != 'both':
nch = 1
ds = PixelMapDataset('training_file_list.txt', pm_ch)
# try out the data loader utility
dl = torch.utils.data.DataLoader(dataset=ds, batch_size=bs, shuffle=True)
# define model
model = None
if modname == 'alexnet':
model = alexnet(num_classes=3, in_ch=nch).to(device)
elif modname == 'densenet':
model = DenseNet(num_classes=3, in_ch=nch).to(device)
elif modname == 'inception':
model = inception_v3(num_classes=3, in_ch=nch).to(device)
elif modname == 'resnet':
model = resnet18(num_classes=3, in_ch=nch).to(device)
elif modname == 'squeezenet':
model = squeezenet1_1(num_classes=3, in_ch=nch).to(device)
elif modname == 'vgg':
model = vgg19_bn(in_ch=nch, num_classes=3).to(device)
else:
print('Model {} not defined.'.format(modname))
return
# loss and optimizer
criterion = nn.CrossEntropyLoss()
optimizer = torch.optim.Adam(model.parameters(), lr=learning_rate)
# training process
total_step = len(dl)
for epoch in range(max_epochs):
for i, (view1, view2, local_labels) in enumerate(dl):
view1 = view1.float().to(device)
if modname == 'inception':
view1 = nn.ZeroPad2d((0, 192, 102, 101))(view1)
else:
view1 = nn.ZeroPad2d((0, 117, 64, 64))(view1)
local_labels = local_labels.to(device)
# forward pass
outputs = model(view1)
loss = criterion(outputs, local_labels)
# backward and optimize
optimizer.zero_grad()
loss.backward()
optimizer.step()
if (i + 1) % bs == 0:
print('Epoch [{}/{}], Step [{}/{}], Loss: {:.4f}'.format(
epoch + 1, max_epochs, i + 1, total_step, loss.item()))
# save the model checkpoint
save_path = '../../../data/two_views/saved_models/{}/{}'.format(
modname, pm_ch)
os.makedirs(save_path, exist_ok=True)
torch.save(model.state_dict(), os.path.join(save_path, 'model.ckpt'))
test = list(test_loader)
test = list(zip(*test))
X_test = torch.cat(test[0], 0)
y_test = torch.cat(test[1], 0)
data = {
'X_train': X_train,
'y_train': y_train,
'X_val': X_val,
'y_val': y_val,
'X_test': X_test,
'y_test': y_test
}
model = DenseNet(input_param=(1, 64),
block_layers=(6, 4),
num_classes=10,
growth_rate=32,
bn_size=2,
dropout_rate=0,
transition_pool_param=(3, 1, 1))
loss_fn = nn.CrossEntropyLoss()
optimizer = torch.optim.Adam(model.parameters(), weight_decay=1e-4)
solver = Solver(model, data, optimizer, loss_fn)
solver.train(num_epoch=2, file_prefix='mnist-')
solver.predict(file_prefix='mnist-')
f"Loaded parameters from saved model: LR is"
f" {helper.params['lr']} and current epoch is {helper.start_epoch}"
)
else:
helper.start_epoch = 1
logger.info(
f'Total number of params for model {helper.params["model"]}: {sum(p.numel() for p in net.parameters() if p.requires_grad)}'
)
if dp:
criterion = nn.CrossEntropyLoss(reduction='none')
else:
criterion = nn.CrossEntropyLoss()
if helper.params['optimizer'] == 'SGD':
optimizer = optim.SGD(net.parameters(),
lr=lr,
momentum=momentum,
weight_decay=decay)
elif helper.params['optimizer'] == 'Adam':
optimizer = optim.Adam(net.parameters(), lr=lr, weight_decay=decay)
else:
raise Exception('Specify `optimizer` in params.yaml.')
scheduler = torch.optim.lr_scheduler.MultiStepLR(
optimizer, milestones=[0.5 * epochs, 0.75 * epochs], gamma=0.1)
table = create_table(helper.params)
writer.add_text('Model Params', table)
logger.info(table)
logger.info(helper.labels)
classes = ('plane', 'car', 'bird', 'cat', 'deer', 'dog', 'frog', 'horse',
'ship', 'truck')
print('==> Making model..')
net = DenseNet(growth_rate=args.growth_rate,
theta=args.theta,
num_layers=[12, 12, 12],
num_classes=10)
net = net.to(device)
if device == 'cuda':
net = torch.nn.DataParallel(net)
cudnn.benchmark = True
num_params = sum(p.numel() for p in net.parameters() if p.requires_grad)
print('The number of parameters of model is', num_params)
if args.resume is not None:
checkpoint = torch.load('./save_model/' + args.resume)
net.load_state_dict(checkpoint['net'])
criterion = nn.CrossEntropyLoss()
optimizer = optim.SGD(net.parameters(),
lr=args.lr,
momentum=args.momentum,
weight_decay=args.weight_decay)
decay_epoch = [150, 225]
scheduler = lr_scheduler.MultiStepLR(optimizer,
milestones=decay_epoch,