def test_net(args):
print("Init...")
_, _, val_loader, _ = lib.build_dataloader(args)
model = lib.build_model(args)
load_model(model, args)
args.cuda = not args.no_cuda and torch.cuda.is_available()
if args.cuda:
model.cuda()
if args.label_smoothing:
criterion = cross_entropy_with_label_smoothing
else:
criterion = nn.CrossEntropyLoss()
print("Start testing...")
val(model, val_loader, criterion, args.test_model, args)
def train_net(args):
print("Init...")
log_writer = tensorboardX.SummaryWriter(args.log_dir)
# train_loader, _, val_loader, _ = lib.build_dataloader(args)
model = lib.build_model(args)
print(model.forward(torch.ones(1, 3, 224, 224)).shape)
# print('Parameters:', sum([np.prod(p.size()) for p in model.parameters()]))
model = torch.nn.DataParallel(model)
optimizer = lib.build_optimizer(args, model)
epoch = 0
if args.resume:
epoch = resume_model(model, optimizer, args)
args.cuda = not args.no_cuda and torch.cuda.is_available()
cudnn.benchmark = True
if args.label_smoothing:
criterion = cross_entropy_with_label_smoothing
else:
# criterion = nn.CrossEntropyLoss()
criterion = nn.MSELoss()
if args.cuda:
model.cuda()
print("Start training...")
while epoch < args.epochs:
train(model, train_loader, optimizer, criterion, epoch, log_writer,
args)
if (epoch + 1) % args.test_epochs == 0:
val(model, val_loader, criterion, epoch, args, log_writer)
if (epoch + 1) % args.save_epochs == 0:
save_model(model, optimizer, epoch, args)
epoch += 1
save_model(model, optimizer, epoch - 1, args)
import tensorflow as tf
from lib import Data_Set, build_model, generate_text
ModelCheckpoint = tf.keras.callbacks.ModelCheckpoint
EPOCHS = 0
data_set = Data_Set(file_path='./texts/websters.txt', verbose=True)
file_name = "./models/websters_model.hdf5"
model = build_model(data_set, file_name=file_name, verbose=True)
checkpoint = ModelCheckpoint(file_name,
monitor='loss',
verbose=1,
save_best_only=True,
mode='min')
def predict():
print(generate_text(build_model, data_set, start_string=u'CRUSTECEAN\n'))
callbacks = [checkpoint]
history = model.fit(data_set.training_data, epochs=EPOCHS, callbacks=callbacks)
temperature = 0.8
num_generate = 500
generated_text = ' '
def train_net(gpu, ngpus_per_node, args):
args.gpu = gpu
if args.gpu is not None:
print("Use GPU: {} for training".format(args.gpu))
if args.distributed:
if args.dist_url == "env://" and args.rank == -1:
args.rank = int(os.environ["RANK"])
if args.multiprocessing_distributed:
# For multiprocessing distributed training, rank needs to be the
# global rank among all the processes
args.rank = args.rank * ngpus_per_node + gpu
dist.init_process_group(backend=args.dist_backend,
init_method=args.dist_url,
world_size=args.world_size,
rank=args.rank)
verbose = args.multiprocessing_distributed and args.rank % ngpus_per_node == 0
if verbose and not args.test:
log_writer = tensorboardX.SummaryWriter(args.log_dir)
else:
log_writer = None
model = lib.build_model(args)
print('Parameters:', sum([np.prod(p.size()) for p in model.parameters()]))
optimizer = lib.build_optimizer(args, model)
if args.distributed:
torch.cuda.set_device(args.gpu)
model.cuda(args.gpu)
args.batch_size = int(args.batch_size / ngpus_per_node)
args.val_batch_size = int(args.val_batch_size / ngpus_per_node)
args.num_workers = int(
(args.num_workers + ngpus_per_node - 1) / ngpus_per_node)
model = torch.nn.parallel.DistributedDataParallel(
model, device_ids=[args.gpu])
epoch = 0
if args.resume:
epoch = resume_model(model, optimizer, args)
if args.label_smoothing:
criterion = cross_entropy_with_label_smoothing
else:
criterion = nn.CrossEntropyLoss()
cudnn.benchmark = True
train_loader, train_sampler, val_loader, val_sampler = lib.build_dataloader(
args, args.distributed)
if args.test:
load_model(model, args)
val(model,
val_loader,
val_sampler,
criterion,
epoch,
verbose=verbose,
args=args)
return
if verbose:
print("Start training...")
while epoch < args.epochs:
train(model,
train_sampler,
train_loader,
optimizer,
criterion,
epoch,
log_writer,
args,
verbose=verbose)
if (epoch + 1) % args.save_epochs == 0:
dist_save_model(model, optimizer, epoch, ngpus_per_node, args)
if (epoch + 1) % args.test_epochs == 0:
val(model,
val_loader,
val_sampler,
criterion,
epoch,
args,
log_writer,
verbose=verbose)
epoch += 1
dist_save_model(model, optimizer, epoch - 1, ngpus_per_node, args)
