def make_lr_fn(base_learning_rate, steps_per_epoch):
if FLAGS.lr_schedule == 'constant':
return lr_schedule.create_constant_learning_rate_schedule(
base_learning_rate, steps_per_epoch)
if FLAGS.lr_schedule == 'stepped':
if not FLAGS.lr_sched_steps:
lr_sched_steps = [[60, 0.2], [120, 0.04], [160, 0.008]]
else:
lr_sched_steps = ast.literal_eval(FLAGS.lr_sched_steps)
return lr_schedule.create_stepped_learning_rate_schedule(
base_learning_rate, steps_per_epoch, lr_sched_steps)
if FLAGS.lr_schedule == 'cosine':
return lr_schedule.create_cosine_learning_rate_schedule(
base_learning_rate, steps_per_epoch, FLAGS.num_epochs)
learning_rate=FLAGS.lr, weight_decay=FLAGS.weight_decay)
elif FLAGS.optimizer == 'Momentum':
optimizer = flax.optim.Momentum(
learning_rate=FLAGS.lr,
beta=FLAGS.momentum,
weight_decay=FLAGS.weight_decay,
nesterov=False)
steps_per_epoch = dataset.get_train_len() // FLAGS.batch_size
if FLAGS.lr_schedule == LR_SCHEDULE_CONSTANT:
lr_fn = lr_schedule.create_constant_learning_rate_schedule(
FLAGS.lr, steps_per_epoch)
elif FLAGS.lr_schedule == LR_SCHEDULE_STEPPED:
lr_schedule_steps = ast.literal_eval(FLAGS.lr_schedule_steps)
lr_fn = lr_schedule.create_stepped_learning_rate_schedule(
FLAGS.lr, steps_per_epoch, lr_schedule_steps)
elif FLAGS.lr_schedule == LR_SCHEDULE_COSINE:
lr_fn = lr_schedule.create_cosine_learning_rate_schedule(
FLAGS.lr, steps_per_epoch, FLAGS.epochs)
else:
raise ValueError(f'Unknown LR schedule type {FLAGS.lr_schedule}')
# Reuses the FLAX learning rate schedule framework for pruning rate schedule.
pruning_fn_p = functools.partial(
lr_schedule.create_stepped_learning_rate_schedule, FLAGS.pruning_rate,
steps_per_epoch)
if FLAGS.pruning_schedule:
pruning_schedule = ast.literal_eval(FLAGS.pruning_schedule)
if isinstance(pruning_schedule, collections.Mapping):
pruning_rate_fn = {
f'MaskedModule_{layer_num}': pruning_fn_p(schedule)
def run_training():
"""Trains a model."""
print('Logging to {}'.format(FLAGS.log_dir))
work_unit_id = uuid.uuid4()
experiment_dir = path.join(FLAGS.experiment_dir, str(work_unit_id))
logging.info('Saving experimental results to %s', experiment_dir)
host_count = jax.host_count()
local_device_count = jax.local_device_count()
logging.info('Device count: %d, host count: %d, local device count: %d',
jax.device_count(), host_count, local_device_count)
if jax.host_id() == 0:
summary_writer = tensorboard.SummaryWriter(experiment_dir)
dataset = dataset_factory.create_dataset(
FLAGS.dataset,
FLAGS.batch_size,
FLAGS.batch_size_test,
shuffle_buffer_size=FLAGS.shuffle_buffer_size)
logging.info('Training %s on the %s dataset...', FLAGS.model,
FLAGS.dataset)
rng = jax.random.PRNGKey(FLAGS.random_seed)
input_shape = (1, ) + dataset.shape
model, initial_state = model_factory.create_model(
FLAGS.model,
rng, ((input_shape, np.float32), ),
num_classes=dataset.num_classes)
if FLAGS.optimizer == 'Adam':
optimizer = flax.optim.Adam(learning_rate=FLAGS.lr,
weight_decay=FLAGS.weight_decay)
elif FLAGS.optimizer == 'Momentum':
optimizer = flax.optim.Momentum(learning_rate=FLAGS.lr,
beta=FLAGS.momentum,
weight_decay=FLAGS.weight_decay,
nesterov=False)
steps_per_epoch = dataset.get_train_len() // FLAGS.batch_size
if FLAGS.lr_schedule == 'constant':
lr_fn = lr_schedule.create_constant_learning_rate_schedule(
FLAGS.lr, steps_per_epoch)
elif FLAGS.lr_schedule == 'stepped':
lr_schedule_steps = ast.literal_eval(FLAGS.lr_schedule_steps)
lr_fn = lr_schedule.create_stepped_learning_rate_schedule(
FLAGS.lr, steps_per_epoch, lr_schedule_steps)
elif FLAGS.lr_schedule == 'cosine':
lr_fn = lr_schedule.create_cosine_learning_rate_schedule(
FLAGS.lr, steps_per_epoch, FLAGS.epochs)
else:
raise ValueError('Unknown LR schedule type {}'.format(
FLAGS.lr_schedule))
if jax.host_id() == 0:
trainer = training.Trainer(
optimizer,
model,
initial_state,
dataset,
rng,
summary_writer=summary_writer,
)
else:
trainer = training.Trainer(optimizer, model, initial_state, dataset,
rng)
_, best_metrics = trainer.train(FLAGS.epochs,
lr_fn=lr_fn,
update_iter=FLAGS.update_iterations,
update_epoch=FLAGS.update_epoch)
logging.info('Best metrics: %s', str(best_metrics))
if jax.host_id() == 0:
for label, value in best_metrics.items():
summary_writer.scalar('best/{}'.format(label), value,
FLAGS.epochs * steps_per_epoch)
summary_writer.close()
def make_clf_lr_fun(base_lr, steps_per_epoch): # pylint: disable=function-redefined
return lr_schedule.create_stepped_learning_rate_schedule(
base_lr, steps_per_epoch, [[60, 0.2], [75, 0.04], [90, 0.008]])
def make_moco_lr_fun(base_lr, steps_per_epoch): # pylint: disable=function-redefined
return lr_schedule.create_stepped_learning_rate_schedule(
base_lr, steps_per_epoch, [[120, 0.1], [160, 0.01]])
def make_lr_fun(base_lr, steps_per_epoch):
return lr_schedule.create_stepped_learning_rate_schedule(
base_lr, steps_per_epoch, lr_sched_steps)
def make_clf_lr_fun(base_lr, steps_per_epoch):
return lr_schedule.create_stepped_learning_rate_schedule(
base_lr,
steps_per_epoch,
lr_clf_sched_steps,
warmup_length=FLAGS.lr_clf_sched_warmup)
def make_lr_fn(base_lr, steps_per_epoch):
return lr_schedule.create_stepped_learning_rate_schedule(
base_lr,
steps_per_epoch,
lr_sched_steps,
warmup_length=config.warmup_epochs)
def make_step_size_fn(steps_per_epoch):
return lr_schedule.create_stepped_learning_rate_schedule(
1.0,
steps_per_epoch,
lr_sched_steps,
warmup_length=config.warmup_epochs)
