def test_interval_is_1(self, mock_save_check):
state = {}
check = Interval('test_file', period=1)
check.on_checkpoint(state)
check.on_checkpoint(state)
self.assertTrue(mock_save_check.call_count == 2)
def test_interval_is_1(self, mock_save_check):
state = {}
check = Interval('test_file', period=1)
check.on_end_epoch(state)
check.on_end_epoch(state)
self.assertTrue(mock_save_check.call_count == 2)
def test_interval_is_more_than_1(self, mock_save_check):
state = {}
check = Interval('test_file', period=4)
for i in range(13):
check.on_checkpoint(state)
if i == 3:
self.assertTrue(mock_save_check.call_count == 1)
elif i == 6:
self.assertFalse(mock_save_check.call_count == 2)
elif i == 7:
self.assertTrue(mock_save_check.call_count == 2)
self.assertTrue(mock_save_check.call_count == 3)
def test_interval_is_more_than_1(self, mock_save_check):
state = {}
check = Interval('test_file', period=4)
for i in range(13):
check.on_end_epoch(state)
if i == 3:
self.assertTrue(mock_save_check.call_count == 1)
elif i == 6:
self.assertFalse(mock_save_check.call_count == 2)
elif i == 7:
self.assertTrue(mock_save_check.call_count == 2)
self.assertTrue(mock_save_check.call_count == 3)
def test_state_dict(self):
check = Interval('test')
check.most_recent = 'temp'
check.epochs_since_last_save = 10
state = check.state_dict()
check = Interval('test')
check.load_state_dict(state)
self.assertEqual(check.most_recent, 'temp')
self.assertEqual(check.epochs_since_last_save, 10)
def train(args,
model,
model_loss,
trainloader,
valloader,
epochs,
name='model'):
init_lr, sched = parse_learning_rate_arg(args.learning_rate)
path = str(args.output)
opt = build_optimiser(args, model, init_lr)
callbacks = [
Interval(filepath=path + '/' + name + '.{epoch:02d}.pt', period=10),
MostRecent(filepath=path + '/' + name + '_final.pt'),
CSVLogger(path + '/' + name + '-train-log.csv'), *sched
]
metrics = ['loss', 'lr']
if isinstance(model_loss, nn.CrossEntropyLoss):
metrics.append('acc')
trial = tb.Trial(model,
opt,
model_loss,
metrics=metrics,
callbacks=callbacks).to(args.device)
trial.with_generators(train_generator=trainloader, val_generator=valloader)
trial.run(epochs=epochs, verbose=2)
return trial
def main():
fake_parser = FakeArgumentParser(add_help=False, allow_abbrev=False)
add_shared_args(fake_parser)
fake_args, _ = fake_parser.parse_known_args()
parser = argparse.ArgumentParser()
add_shared_args(parser)
add_sub_args(fake_args, parser)
args = parser.parse_args()
trainloader, valloader, testloader = build_dataloaders(args)
args.output.mkdir(exist_ok=True, parents=True)
path = str(args.output)
save_args(args.output)
model = get_model(args.model)()
init_lr, sched = parse_learning_rate_arg(args.learning_rate)
if args.optimiser == 'Adam':
opt = optim.Adam(model.parameters(),
lr=init_lr,
weight_decay=args.weight_decay)
else:
opt = optim.SGD(model.parameters(),
lr=init_lr,
weight_decay=args.weight_decay,
momentum=args.momentum)
callbacks = [
Interval(filepath=path + '/model.{epoch:02d}.pt', period=10),
MostRecent(filepath=path + '/model_final.pt'),
CSVLogger(path + '/train-log.csv'), *sched
]
trial = tb.Trial(model,
opt,
torch.nn.CrossEntropyLoss(),
metrics=['loss', 'acc', 'lr'],
callbacks=callbacks).to(args.device)
trial.with_generators(train_generator=trainloader, val_generator=valloader)
trial.run(epochs=args.epochs, verbose=2)
trial = tb.Trial(model,
criterion=torch.nn.CrossEntropyLoss(),
metrics=['loss', 'acc'],
callbacks=[CSVLogger(path + '/test-log.csv')
]).to(args.device)
trial.with_generators(test_generator=testloader)
trial.predict(verbose=2)
def test_interval_on_batch(self, mock_save_check):
state = {}
check = Interval('test_file', period=4, on_batch=True)
for i in range(13):
check.on_step_training(state)
if i == 3:
self.assertTrue(mock_save_check.call_count == 1)
elif i == 6:
self.assertFalse(mock_save_check.call_count == 2)
elif i == 7:
self.assertTrue(mock_save_check.call_count == 2)
check.on_checkpoint(state)
self.assertTrue(mock_save_check.call_count == 3)
