def run(options):
if options.device is not None:
pt_training.set_device(options.device)
device = get_device()
pt_training.set_random_seed(options.seed)
folder = join(config.EXP_FOLDER, options.folder)
if not options.do_overwrite:
if abort_due_to_overwrite_safety(folder):
print('Process aborted.')
return
check_mkdir(folder)
if options.comment is None:
print(
'You forgot to add a comment to your experiment. Please add something!'
)
options.comment = input('Comment: ')
save_options(join(folder, 'opt.json'), options)
copy_source(folder, do_not_copy_folders=config.DO_NOT_COPY)
_train_model(options, folder, device)
def save_np(data, labels, path):
filename = f'{path}.npy'
check_mkdir(filename)
np.save(filename, data)
print(f'saving to {filename} finished')
filename = f'{path}_labels.npy'
check_mkdir(filename)
np.save(filename, labels)
print(f'saving to {filename} finished')
def create_splits(num_splits=5, split_perc=.8):
images_ = _get_images_sorted_by_class()
for i in range(num_splits):
train_images = []
test_images = []
# for each class get split_perc % for training
for tmp in images_.values():
n = len(tmp)
n_train = int(split_perc * n)
# grab images without replacement
tmp_train = random.sample(tmp, n_train)
tmp_test = list(set(tmp) - set(tmp_train))
train_images += tmp_train
test_images += tmp_test
out_path = join(config.NAT_IM_BASE, 'splits', f'{i}.json')
check_mkdir(out_path)
with open(out_path, 'w') as file:
json.dump({'train': train_images, 'test': test_images}, file)
def _train_model(options, folder, device):
model_out = join(folder, 'model.pt')
log_file = join(folder, 'log.json')
check_mkdir(log_file)
model = get_model(options.model_type, options.in_dim, options.out_dim,
device)
if options.model_path is not None:
model_sd = torch.load(options.model_path).state_dict()
model.load_state_dict(model_sd, strict=False)
write_model_specs(folder, model)
optimizer = pt_training.get_optimizer(options.opt,
model.parameters(),
options.lr,
momentum=options.mom,
weight_decay=options.wd)
if options.lr_steps > 0 or options.fixed_steps is not None:
scheduler = pt_training.get_scheduler(options.lr_steps,
options.epochs,
optimizer,
fixed_steps=options.fixed_steps)
else:
print('no scheduling used')
scheduler = None
data_loaders = get_data_loaders(options.dataset,
options.bs,
split_index=options.split_index)
if options.early_stopping:
assert options.sv_int == -1
early_stopping = pt_training.EarlyStopping(options.es_metric,
get_max=True,
epoch_thres=options.epochs)
else:
early_stopping = None
train_interface = get_interface(
'classification',
model,
device,
Printer(config.PRINT_FREQUENCY, log_file),
)
training = pt_training.Training(optimizer,
data_loaders['train'],
train_interface,
scheduler=scheduler,
printer=train_interface.printer,
save_path=model_out,
save_steps=options.sv_int,
val_data_loader=data_loaders['val'],
early_stopping=early_stopping,
save_state_dict=True,
test_data_loader=data_loaders['test'])
training(options.epochs)