def test_ofa(self):
ofa_model = OFA(self.model,
self.run_config,
distill_config=self.distill_config,
elastic_order=self.elastic_order)
start_epoch = 0
for idx in range(len(self.run_config.n_epochs)):
cur_idx = self.run_config.n_epochs[idx]
for ph_idx in range(len(cur_idx)):
cur_lr = self.run_config.init_learning_rate[idx][ph_idx]
adam = paddle.optimizer.Adam(
learning_rate=cur_lr,
parameters=(ofa_model.parameters() + ofa_model.netAs_param))
for epoch_id in range(start_epoch,
self.run_config.n_epochs[idx][ph_idx]):
if epoch_id == 0:
ofa_model.set_epoch(epoch_id)
for model_no in range(self.run_config.dynamic_batch_size[
idx]):
output, _ = ofa_model(self.data)
loss = paddle.mean(output)
if self.distill_config.mapping_layers != None:
dis_loss = ofa_model.calc_distill_loss()
loss += dis_loss
dis_loss = dis_loss.numpy()[0]
else:
dis_loss = 0
print('epoch: {}, loss: {}, distill loss: {}'.format(
epoch_id, loss.numpy()[0], dis_loss))
loss.backward()
adam.minimize(loss)
adam.clear_gradients()
start_epoch = self.run_config.n_epochs[idx][ph_idx]
class TestOFAV2(unittest.TestCase):
def setUp(self):
model = ModelV1()
sp_net_config = supernet(expand_ratio=[0.25, 0.5, 1.0])
self.model = Convert(sp_net_config).convert(model)
self.images = paddle.randn(shape=[2, 3, 32, 32], dtype='float32')
def test_ofa(self):
self.ofa_model = OFA(self.model)
self.ofa_model.set_epoch(0)
self.ofa_model.set_task('expand_ratio')
out, _ = self.ofa_model(self.images)
class TestExportCase1(unittest.TestCase):
def setUp(self):
model = ModelLinear1()
data_np = np.random.random((3, 64)).astype(np.int64)
self.data = paddle.to_tensor(data_np)
self.ofa_model = OFA(model)
self.ofa_model.set_epoch(0)
outs, _ = self.ofa_model(self.data)
self.config = self.ofa_model.current_config
def test_export_model(self):
self.ofa_model.export(
self.config, input_shapes=[[3, 64]], input_dtypes=['int64'])
assert len(self.ofa_model.ofa_layers) == 4
class Testelementwise(unittest.TestCase):
def setUp(self):
model = ModelElementwise()
sp_net_config = supernet(expand_ratio=[0.25, 0.5, 1.0])
self.model = Convert(sp_net_config).convert(model)
self.images = paddle.randn(shape=[2, 3, 32, 32], dtype='float32')
def test_elementwise(self):
self.ofa_model = OFA(self.model)
self.ofa_model.set_epoch(0)
self.ofa_model.set_task('expand_ratio')
out, _ = self.ofa_model(self.images)
assert list(
self.ofa_model._ofa_layers.keys()) == ['conv2.0', 'conv3.0']
class TestMultiExit(unittest.TestCase):
def setUp(self):
self.images = paddle.randn(shape=[1, 3, 224, 224], dtype='float32')
model = ModelMultiExit()
sp_net_config = supernet(expand_ratio=[0.25, 0.5, 1.0])
self.model = Convert(sp_net_config).convert(model)
def test_multiexit(self):
self.ofa_model = OFA(self.model)
self.ofa_model.set_epoch(0)
self.ofa_model.set_task('expand_ratio')
out, _ = self.ofa_model(self.images)
assert list(self.ofa_model._ofa_layers.keys()) == [
'conv1.0', 'block1.0', 'block1.4', 'block2.0', 'out2.0'
]
class TestExportCase2(unittest.TestCase):
def setUp(self):
model = ModelLinear()
data_np = np.random.random((3, 64)).astype(np.int64)
self.data = paddle.to_tensor(data_np)
self.ofa_model = OFA(model)
self.ofa_model.set_epoch(0)
outs, _ = self.ofa_model(self.data)
self.config = self.ofa_model.current_config
def test_export_model_linear2(self):
config = self.ofa_model._sample_config(
task='expand_ratio', phase=None, sample_type='smallest')
ex_model = self.ofa_model.export(
config, input_shapes=[[3, 64]], input_dtypes=['int64'])
ex_model(self.data)
assert len(self.ofa_model.ofa_layers) == 3
class TestShortCut(unittest.TestCase):
def setUp(self):
model = resnet50()
sp_net_config = supernet(expand_ratio=[0.25, 0.5, 1.0])
self.model = Convert(sp_net_config).convert(model)
self.images = paddle.randn(shape=[2, 3, 224, 224], dtype='float32')
self._test_clear_search_space()
def _test_clear_search_space(self):
self.ofa_model = OFA(self.model)
self.ofa_model.set_epoch(0)
outs, _ = self.ofa_model(self.images)
self.config = self.ofa_model.current_config
def test_export_model(self):
self.ofa_model.export(
self.config,
input_shapes=[[2, 3, 224, 224]],
input_dtypes=['float32'])
assert len(self.ofa_model.ofa_layers) == 37
class TestShortcutSkiplayers(unittest.TestCase):
def setUp(self):
model = ModelShortcut()
sp_net_config = supernet(expand_ratio=[0.5, 1.0])
self.model = Convert(sp_net_config).convert(model)
self.images = paddle.randn(shape=[2, 3, 32, 32], dtype='float32')
self.init_config()
self.ofa_model = OFA(self.model, run_config=self.run_config)
self.ofa_model._clear_search_space(self.images)
def init_config(self):
default_run_config = {'skip_layers': ['branch1.6']}
self.run_config = RunConfig(**default_run_config)
def test_shortcut(self):
self.ofa_model.set_epoch(0)
self.ofa_model.set_task('expand_ratio')
for i in range(5):
self.ofa_model(self.images)
assert list(self.ofa_model._ofa_layers.keys()) == ['branch2.0']
def do_train(args):
paddle.set_device("gpu" if args.n_gpu else "cpu")
if paddle.distributed.get_world_size() > 1:
paddle.distributed.init_parallel_env()
set_seed(args)
args.task_name = args.task_name.lower()
dataset_class, metric_class = TASK_CLASSES[args.task_name]
args.model_type = args.model_type.lower()
model_class, tokenizer_class = MODEL_CLASSES[args.model_type]
train_ds = dataset_class.get_datasets(['train'])
tokenizer = tokenizer_class.from_pretrained(args.model_name_or_path)
trans_func = partial(convert_example,
tokenizer=tokenizer,
label_list=train_ds.get_labels(),
max_seq_length=args.max_seq_length)
train_ds = train_ds.apply(trans_func, lazy=True)
train_batch_sampler = paddle.io.DistributedBatchSampler(
train_ds, batch_size=args.batch_size, shuffle=True)
batchify_fn = lambda samples, fn=Tuple(
Pad(axis=0, pad_val=tokenizer.pad_token_id), # input
Pad(axis=0, pad_val=tokenizer.pad_token_id), # segment
Stack(), # length
Stack(dtype="int64" if train_ds.get_labels() else "float32") # label
): [data for i, data in enumerate(fn(samples)) if i != 2]
train_data_loader = DataLoader(dataset=train_ds,
batch_sampler=train_batch_sampler,
collate_fn=batchify_fn,
num_workers=0,
return_list=True)
if args.task_name == "mnli":
dev_dataset_matched, dev_dataset_mismatched = dataset_class.get_datasets(
["dev_matched", "dev_mismatched"])
dev_dataset_matched = dev_dataset_matched.apply(trans_func, lazy=True)
dev_dataset_mismatched = dev_dataset_mismatched.apply(trans_func,
lazy=True)
dev_batch_sampler_matched = paddle.io.BatchSampler(
dev_dataset_matched, batch_size=args.batch_size, shuffle=False)
dev_data_loader_matched = DataLoader(
dataset=dev_dataset_matched,
batch_sampler=dev_batch_sampler_matched,
collate_fn=batchify_fn,
num_workers=0,
return_list=True)
dev_batch_sampler_mismatched = paddle.io.BatchSampler(
dev_dataset_mismatched, batch_size=args.batch_size, shuffle=False)
dev_data_loader_mismatched = DataLoader(
dataset=dev_dataset_mismatched,
batch_sampler=dev_batch_sampler_mismatched,
collate_fn=batchify_fn,
num_workers=0,
return_list=True)
else:
dev_dataset = dataset_class.get_datasets(["dev"])
dev_dataset = dev_dataset.apply(trans_func, lazy=True)
dev_batch_sampler = paddle.io.BatchSampler(dev_dataset,
batch_size=args.batch_size,
shuffle=False)
dev_data_loader = DataLoader(dataset=dev_dataset,
batch_sampler=dev_batch_sampler,
collate_fn=batchify_fn,
num_workers=0,
return_list=True)
num_labels = 1 if train_ds.get_labels() == None else len(
train_ds.get_labels())
model = model_class.from_pretrained(args.model_name_or_path,
num_classes=num_labels)
if paddle.distributed.get_world_size() > 1:
model = paddle.DataParallel(model)
# Step1: Initialize a dictionary to save the weights from the origin BERT model.
origin_weights = {}
for name, param in model.named_parameters():
origin_weights[name] = param
# Step2: Convert origin model to supernet.
sp_config = supernet(expand_ratio=args.width_mult_list)
model = Convert(sp_config).convert(model)
# Use weights saved in the dictionary to initialize supernet.
utils.set_state_dict(model, origin_weights)
del origin_weights
# Step3: Define teacher model.
teacher_model = model_class.from_pretrained(args.model_name_or_path,
num_classes=num_labels)
# Step4: Config about distillation.
mapping_layers = ['bert.embeddings']
for idx in range(model.bert.config['num_hidden_layers']):
mapping_layers.append('bert.encoder.layers.{}'.format(idx))
default_distill_config = {
'lambda_distill': 0.1,
'teacher_model': teacher_model,
'mapping_layers': mapping_layers,
}
distill_config = DistillConfig(**default_distill_config)
# Step5: Config in supernet training.
ofa_model = OFA(model,
distill_config=distill_config,
elastic_order=['width'])
criterion = paddle.nn.loss.CrossEntropyLoss() if train_ds.get_labels(
) else paddle.nn.loss.MSELoss()
metric = metric_class()
if args.task_name == "mnli":
dev_data_loader = (dev_data_loader_matched, dev_data_loader_mismatched)
# Step6: Calculate the importance of neurons and head,
# and then reorder them according to the importance.
head_importance, neuron_importance = utils.compute_neuron_head_importance(
args.task_name,
ofa_model.model,
dev_data_loader,
loss_fct=criterion,
num_layers=model.bert.config['num_hidden_layers'],
num_heads=model.bert.config['num_attention_heads'])
reorder_neuron_head(ofa_model.model, head_importance, neuron_importance)
lr_scheduler = paddle.optimizer.lr.LambdaDecay(
args.learning_rate,
lambda current_step, num_warmup_steps=args.warmup_steps,
num_training_steps=args.max_steps if args.max_steps > 0 else
(len(train_data_loader) * args.num_train_epochs): float(
current_step) / float(max(1, num_warmup_steps))
if current_step < num_warmup_steps else max(
0.0,
float(num_training_steps - current_step) / float(
max(1, num_training_steps - num_warmup_steps))))
optimizer = paddle.optimizer.AdamW(
learning_rate=lr_scheduler,
epsilon=args.adam_epsilon,
parameters=ofa_model.model.parameters(),
weight_decay=args.weight_decay,
apply_decay_param_fun=lambda x: x in [
p.name for n, p in ofa_model.model.named_parameters()
if not any(nd in n for nd in ["bias", "norm"])
])
global_step = 0
tic_train = time.time()
for epoch in range(args.num_train_epochs):
# Step7: Set current epoch and task.
ofa_model.set_epoch(epoch)
ofa_model.set_task('width')
for step, batch in enumerate(train_data_loader):
global_step += 1
input_ids, segment_ids, labels = batch
for width_mult in args.width_mult_list:
# Step8: Broadcast supernet config from width_mult,
# and use this config in supernet training.
net_config = apply_config(ofa_model, width_mult)
ofa_model.set_net_config(net_config)
logits, teacher_logits = ofa_model(input_ids,
segment_ids,
attention_mask=[None, None])
rep_loss = ofa_model.calc_distill_loss()
if args.task_name == 'sts-b':
logit_loss = 0.0
else:
logit_loss = soft_cross_entropy(logits,
teacher_logits.detach())
loss = rep_loss + args.lambda_logit * logit_loss
loss.backward()
optimizer.step()
lr_scheduler.step()
ofa_model.model.clear_gradients()
if global_step % args.logging_steps == 0:
if (not args.n_gpu > 1) or paddle.distributed.get_rank() == 0:
logger.info(
"global step %d, epoch: %d, batch: %d, loss: %f, speed: %.2f step/s"
% (global_step, epoch, step, loss, args.logging_steps /
(time.time() - tic_train)))
tic_train = time.time()
if global_step % args.save_steps == 0:
if args.task_name == "mnli":
evaluate(teacher_model,
criterion,
metric,
dev_data_loader_matched,
width_mult=100)
evaluate(teacher_model,
criterion,
metric,
dev_data_loader_mismatched,
width_mult=100)
else:
evaluate(teacher_model,
criterion,
metric,
dev_data_loader,
width_mult=100)
for idx, width_mult in enumerate(args.width_mult_list):
net_config = apply_config(ofa_model, width_mult)
ofa_model.set_net_config(net_config)
tic_eval = time.time()
if args.task_name == "mnli":
acc = evaluate(ofa_model, criterion, metric,
dev_data_loader_matched, width_mult)
evaluate(ofa_model, criterion, metric,
dev_data_loader_mismatched, width_mult)
print("eval done total : %s s" %
(time.time() - tic_eval))
else:
acc = evaluate(ofa_model, criterion, metric,
dev_data_loader, width_mult)
print("eval done total : %s s" %
(time.time() - tic_eval))
if (not args.n_gpu > 1
) or paddle.distributed.get_rank() == 0:
output_dir = os.path.join(args.output_dir,
"model_%d" % global_step)
if not os.path.exists(output_dir):
os.makedirs(output_dir)
# need better way to get inner model of DataParallel
model_to_save = model._layers if isinstance(
model, paddle.DataParallel) else model
model_to_save.save_pretrained(output_dir)
tokenizer.save_pretrained(output_dir)
g_clip = F.clip.GradientClipByGlobalNorm(1.0) #experimental
if args.use_lr_decay:
opt = AdamW(learning_rate=LinearDecay(
args.lr, int(args.warmup_proportion * args.max_steps),
args.max_steps),
parameter_list=ofa_model.model.parameters(),
weight_decay=args.wd,
grad_clip=g_clip)
else:
opt = AdamW(args.lr,
parameter_list=ofa_model.model.parameters(),
weight_decay=args.wd,
grad_clip=g_clip)
for epoch in range(max(run_config.n_epochs[-1])):
ofa_model.set_epoch(epoch)
if epoch <= int(max(run_config.n_epochs[0])):
ofa_model.set_task('width')
depth_mult_list = [1.0]
else:
ofa_model.set_task('depth')
depth_mult_list = run_config.elastic_depth
for step, d in enumerate(
tqdm(train_ds.start(place), desc='training')):
ids, sids, label = d
accumulate_gradients = dict()
for param in opt._parameter_list:
accumulate_gradients[param.name] = 0.0
for depth_mult in depth_mult_list:
def do_train(args):
paddle.set_device(args.device)
if paddle.distributed.get_world_size() > 1:
paddle.distributed.init_parallel_env()
set_seed(args)
args.task_name = args.task_name.lower()
metric_class = METRIC_CLASSES[args.task_name]
args.model_type = args.model_type.lower()
model_class, tokenizer_class = MODEL_CLASSES[args.model_type]
train_ds = load_dataset('clue', args.task_name, splits='train')
tokenizer = tokenizer_class.from_pretrained(args.model_name_or_path)
trans_func = partial(convert_example,
label_list=train_ds.label_list,
tokenizer=tokenizer,
max_seq_length=args.max_seq_length)
train_ds = train_ds.map(trans_func, lazy=True)
train_batch_sampler = paddle.io.DistributedBatchSampler(
train_ds, batch_size=args.batch_size, shuffle=True)
batchify_fn = lambda samples, fn=Tuple(
Pad(axis=0, pad_val=tokenizer.pad_token_id), # input
Pad(axis=0, pad_val=tokenizer.pad_token_type_id), # segment
Stack(dtype="int64" if train_ds.label_list else "float32") # label
): fn(samples)
train_data_loader = DataLoader(dataset=train_ds,
batch_sampler=train_batch_sampler,
collate_fn=batchify_fn,
num_workers=0,
return_list=True)
dev_ds = load_dataset('clue', args.task_name, splits='dev')
dev_ds = dev_ds.map(trans_func, lazy=True)
dev_batch_sampler = paddle.io.BatchSampler(dev_ds,
batch_size=args.batch_size,
shuffle=False)
dev_data_loader = DataLoader(dataset=dev_ds,
batch_sampler=dev_batch_sampler,
collate_fn=batchify_fn,
num_workers=0,
return_list=True)
num_labels = 1 if train_ds.label_list == None else len(train_ds.label_list)
model = model_class.from_pretrained(args.model_name_or_path,
num_classes=num_labels)
# Step1: Initialize a dictionary to save the weights from the origin PPMiniLM model.
origin_weights = model.state_dict()
# Step2: Convert origin model to supernet.
sp_config = supernet(expand_ratio=[1.0])
model = Convert(sp_config).convert(model)
# Use weights saved in the dictionary to initialize supernet.
utils.set_state_dict(model, origin_weights)
del origin_weights
super_sd = paddle.load(
os.path.join(args.model_name_or_path, 'model_state.pdparams'))
model.set_state_dict(super_sd)
# Step3: Define teacher model.
teacher_model = model_class.from_pretrained(args.model_name_or_path,
num_classes=num_labels)
# Step4: Config about distillation.
mapping_layers = ['ppminilm.embeddings']
for idx in range(model.ppminilm.config['num_hidden_layers']):
mapping_layers.append('ppminilm.encoder.layers.{}'.format(idx))
default_distill_config = {
'lambda_distill': 0.1,
'teacher_model': teacher_model,
'mapping_layers': mapping_layers,
}
distill_config = DistillConfig(**default_distill_config)
# Step5: Config in supernet training.
ofa_model = OFA(model,
distill_config=distill_config,
elastic_order=['width'])
criterion = paddle.nn.loss.CrossEntropyLoss(
) if train_ds.label_list else paddle.nn.loss.MSELoss()
metric = metric_class()
#### Step6: Calculate the importance of neurons and head,
#### and then reorder them according to the importance.
head_importance, neuron_importance = nlp_utils.compute_neuron_head_importance(
args.task_name,
ofa_model.model,
dev_data_loader,
loss_fct=criterion,
num_layers=model.ppminilm.config['num_hidden_layers'],
num_heads=model.ppminilm.config['num_attention_heads'])
reorder_neuron_head(ofa_model.model, head_importance, neuron_importance)
if paddle.distributed.get_world_size() > 1:
ofa_model.model = paddle.DataParallel(ofa_model.model)
if args.max_steps > 0:
num_training_steps = args.max_steps
num_train_epochs = math.ceil(num_training_steps /
len(train_data_loader))
else:
num_training_steps = len(train_data_loader) * args.num_train_epochs
num_train_epochs = args.num_train_epochs
warmup = args.warmup_steps if args.warmup_steps > 0 else args.warmup_proportion
lr_scheduler = LinearDecayWithWarmup(args.learning_rate,
num_training_steps, warmup)
# Generate parameter names needed to perform weight decay.
# All bias and LayerNorm parameters are excluded.
decay_params = [
p.name for n, p in model.named_parameters()
if not any(nd in n for nd in ["bias", "norm"])
]
optimizer = paddle.optimizer.AdamW(
learning_rate=lr_scheduler,
beta1=0.9,
beta2=0.999,
epsilon=args.adam_epsilon,
parameters=model.parameters(),
weight_decay=args.weight_decay,
apply_decay_param_fun=lambda x: x in decay_params,
grad_clip=nn.ClipGradByGlobalNorm(args.max_grad_norm))
global_step = 0
tic_train = time.time()
best_res = 0.0
for epoch in range(num_train_epochs):
# Step7: Set current epoch and task.
ofa_model.set_epoch(epoch)
ofa_model.set_task('width')
for step, batch in enumerate(train_data_loader):
global_step += 1
input_ids, segment_ids, labels = batch
for width_mult in args.width_mult_list:
# Step8: Broadcast supernet config from width_mult,
# and use this config in supernet training.
net_config = utils.dynabert_config(ofa_model, width_mult)
ofa_model.set_net_config(net_config)
logits, teacher_logits = ofa_model(input_ids,
segment_ids,
attention_mask=[None, None])
rep_loss = ofa_model.calc_distill_loss()
logit_loss = soft_cross_entropy(logits,
teacher_logits.detach())
loss = rep_loss + args.lambda_logit * logit_loss
loss.backward()
optimizer.step()
lr_scheduler.step()
optimizer.clear_grad()
if global_step % args.logging_steps == 0:
logger.info(
"global step %d, epoch: %d, batch: %d, loss: %f, speed: %.2f step/s"
% (global_step, epoch, step, loss, args.logging_steps /
(time.time() - tic_train)))
tic_train = time.time()
if global_step % args.save_steps == 0 or global_step == num_training_steps:
tic_eval = time.time()
evaluate(teacher_model,
metric,
dev_data_loader,
width_mult=100)
print("eval done total : %s s" % (time.time() - tic_eval))
for idx, width_mult in enumerate(args.width_mult_list):
net_config = utils.dynabert_config(ofa_model, width_mult)
ofa_model.set_net_config(net_config)
tic_eval = time.time()
res = evaluate(ofa_model, metric, dev_data_loader,
width_mult)
print("eval done total : %s s" % (time.time() - tic_eval))
if best_res < res:
output_dir = args.output_dir
if not os.path.exists(output_dir):
os.makedirs(output_dir)
# need better way to get inner model of DataParallel
model_to_save = model._layers if isinstance(
model, paddle.DataParallel) else model
model_to_save.save_pretrained(output_dir)
tokenizer.save_pretrained(output_dir)
best_res = res
if global_step >= num_training_steps:
print("best_res: ", best_res)
return
print("best_res: ", best_res)
def do_train(args):
paddle.set_device(args.device)
if paddle.distributed.get_world_size() > 1:
paddle.distributed.init_parallel_env()
set_seed(args)
args.task_name = args.task_name.lower()
metric_class = METRIC_CLASSES[args.task_name]
args.model_type = args.model_type.lower()
model_class, tokenizer_class = MODEL_CLASSES[args.model_type]
train_ds = load_dataset('glue', args.task_name, splits="train")
tokenizer = tokenizer_class.from_pretrained(args.model_name_or_path)
trans_func = partial(
convert_example,
tokenizer=tokenizer,
label_list=train_ds.label_list,
max_seq_length=args.max_seq_length)
train_ds = train_ds.map(trans_func, lazy=True)
train_batch_sampler = paddle.io.DistributedBatchSampler(
train_ds, batch_size=args.batch_size, shuffle=True)
batchify_fn = lambda samples, fn=Tuple(
Pad(axis=0, pad_val=tokenizer.pad_token_id), # input
Pad(axis=0, pad_val=tokenizer.pad_token_type_id), # segment
Stack(dtype="int64" if train_ds.label_list else "float32") # label
): fn(samples)
train_data_loader = DataLoader(
dataset=train_ds,
batch_sampler=train_batch_sampler,
collate_fn=batchify_fn,
num_workers=0,
return_list=True)
if args.task_name == "mnli":
dev_ds_matched, dev_ds_mismatched = load_dataset(
'glue', args.task_name, splits=["dev_matched", "dev_mismatched"])
dev_ds_matched = dev_ds_matched.map(trans_func, lazy=True)
dev_ds_mismatched = dev_ds_mismatched.map(trans_func, lazy=True)
dev_batch_sampler_matched = paddle.io.BatchSampler(
dev_ds_matched, batch_size=args.batch_size, shuffle=False)
dev_data_loader_matched = DataLoader(
dataset=dev_ds_matched,
batch_sampler=dev_batch_sampler_matched,
collate_fn=batchify_fn,
num_workers=0,
return_list=True)
dev_batch_sampler_mismatched = paddle.io.BatchSampler(
dev_ds_mismatched, batch_size=args.batch_size, shuffle=False)
dev_data_loader_mismatched = DataLoader(
dataset=dev_ds_mismatched,
batch_sampler=dev_batch_sampler_mismatched,
collate_fn=batchify_fn,
num_workers=0,
return_list=True)
else:
dev_ds = load_dataset('glue', args.task_name, splits='dev')
dev_ds = dev_ds.map(trans_func, lazy=True)
dev_batch_sampler = paddle.io.BatchSampler(
dev_ds, batch_size=args.batch_size, shuffle=False)
dev_data_loader = DataLoader(
dataset=dev_ds,
batch_sampler=dev_batch_sampler,
collate_fn=batchify_fn,
num_workers=0,
return_list=True)
num_labels = 1 if train_ds.label_list == None else len(train_ds.label_list)
# Step1: Initialize the origin BERT model.
model = model_class.from_pretrained(
args.model_name_or_path, num_classes=num_labels)
origin_weights = model.state_dict()
# Step2: Convert origin model to supernet.
sp_config = supernet(expand_ratio=args.width_mult_list)
model = Convert(sp_config).convert(model)
# Use weights saved in the dictionary to initialize supernet.
utils.set_state_dict(model, origin_weights)
# Step3: Define teacher model.
teacher_model = model_class.from_pretrained(
args.model_name_or_path, num_classes=num_labels)
new_dict = utils.utils.remove_model_fn(teacher_model, origin_weights)
teacher_model.set_state_dict(new_dict)
del origin_weights, new_dict
default_run_config = {'elastic_depth': args.depth_mult_list}
run_config = RunConfig(**default_run_config)
# Step4: Config about distillation.
mapping_layers = ['bert.embeddings']
for idx in range(model.bert.config['num_hidden_layers']):
mapping_layers.append('bert.encoder.layers.{}'.format(idx))
default_distill_config = {
'lambda_distill': args.lambda_rep,
'teacher_model': teacher_model,
'mapping_layers': mapping_layers,
}
distill_config = DistillConfig(**default_distill_config)
# Step5: Config in supernet training.
ofa_model = OFA(model,
run_config=run_config,
distill_config=distill_config,
elastic_order=['depth'])
#elastic_order=['width'])
criterion = paddle.nn.CrossEntropyLoss(
) if train_ds.label_list else paddle.nn.MSELoss()
metric = metric_class()
if args.task_name == "mnli":
dev_data_loader = (dev_data_loader_matched, dev_data_loader_mismatched)
if paddle.distributed.get_world_size() > 1:
ofa_model.model = paddle.DataParallel(
ofa_model.model, find_unused_parameters=True)
if args.max_steps > 0:
num_training_steps = args.max_steps
num_train_epochs = math.ceil(num_training_steps /
len(train_data_loader))
else:
num_training_steps = len(train_data_loader) * args.num_train_epochs
num_train_epochs = args.num_train_epochs
lr_scheduler = LinearDecayWithWarmup(args.learning_rate, num_training_steps,
args.warmup_steps)
# Generate parameter names needed to perform weight decay.
# All bias and LayerNorm parameters are excluded.
decay_params = [
p.name for n, p in model.named_parameters()
if not any(nd in n for nd in ["bias", "norm"])
]
optimizer = paddle.optimizer.AdamW(
learning_rate=lr_scheduler,
epsilon=args.adam_epsilon,
parameters=ofa_model.model.parameters(),
weight_decay=args.weight_decay,
apply_decay_param_fun=lambda x: x in decay_params)
global_step = 0
tic_train = time.time()
for epoch in range(num_train_epochs):
# Step6: Set current epoch and task.
ofa_model.set_epoch(epoch)
ofa_model.set_task('depth')
for step, batch in enumerate(train_data_loader):
global_step += 1
input_ids, segment_ids, labels = batch
for depth_mult in args.depth_mult_list:
for width_mult in args.width_mult_list:
# Step7: Broadcast supernet config from width_mult,
# and use this config in supernet training.
net_config = utils.dynabert_config(ofa_model, width_mult,
depth_mult)
ofa_model.set_net_config(net_config)
logits, teacher_logits = ofa_model(
input_ids, segment_ids, attention_mask=[None, None])
rep_loss = ofa_model.calc_distill_loss()
if args.task_name == 'sts-b':
logit_loss = 0.0
else:
logit_loss = soft_cross_entropy(logits,
teacher_logits.detach())
loss = rep_loss + args.lambda_logit * logit_loss
loss.backward()
optimizer.step()
lr_scheduler.step()
ofa_model.model.clear_gradients()
if global_step % args.logging_steps == 0:
if paddle.distributed.get_rank() == 0:
logger.info(
"global step %d, epoch: %d, batch: %d, loss: %f, speed: %.2f step/s"
% (global_step, epoch, step, loss,
args.logging_steps / (time.time() - tic_train)))
tic_train = time.time()
if global_step % args.save_steps == 0:
if args.task_name == "mnli":
evaluate(
teacher_model,
criterion,
metric,
dev_data_loader_matched,
width_mult=100)
evaluate(
teacher_model,
criterion,
metric,
dev_data_loader_mismatched,
width_mult=100)
else:
evaluate(
teacher_model,
criterion,
metric,
dev_data_loader,
width_mult=100)
for depth_mult in args.depth_mult_list:
for width_mult in args.width_mult_list:
net_config = utils.dynabert_config(
ofa_model, width_mult, depth_mult)
ofa_model.set_net_config(net_config)
tic_eval = time.time()
if args.task_name == "mnli":
acc = evaluate(ofa_model, criterion, metric,
dev_data_loader_matched, width_mult,
depth_mult)
evaluate(ofa_model, criterion, metric,
dev_data_loader_mismatched, width_mult,
depth_mult)
print("eval done total : %s s" %
(time.time() - tic_eval))
else:
acc = evaluate(ofa_model, criterion, metric,
dev_data_loader, width_mult,
depth_mult)
print("eval done total : %s s" %
(time.time() - tic_eval))
if paddle.distributed.get_rank() == 0:
output_dir = os.path.join(args.output_dir,
"model_%d" % global_step)
if not os.path.exists(output_dir):
os.makedirs(output_dir)
# need better way to get inner model of DataParallel
model_to_save = model._layers if isinstance(
model, paddle.DataParallel) else model
model_to_save.save_pretrained(output_dir)
tokenizer.save_pretrained(output_dir)
if global_step >= num_training_steps:
return
def __call__(self, model, param_state_dict):
paddleslim = try_import('paddleslim')
from paddleslim.nas.ofa import OFA, RunConfig, utils
from paddleslim.nas.ofa.convert_super import Convert, supernet
task = self.ofa_config['task']
expand_ratio = self.ofa_config['expand_ratio']
skip_neck = self.ofa_config['skip_neck']
skip_head = self.ofa_config['skip_head']
run_config = self.ofa_config['RunConfig']
if 'skip_layers' in run_config:
skip_layers = run_config['skip_layers']
else:
skip_layers = []
# supernet config
sp_config = supernet(expand_ratio=expand_ratio)
# convert to supernet
model = Convert(sp_config).convert(model)
skip_names = []
if skip_neck:
skip_names.append('neck.')
if skip_head:
skip_names.append('head.')
for name, sublayer in model.named_sublayers():
for n in skip_names:
if n in name:
skip_layers.append(name)
run_config['skip_layers'] = skip_layers
run_config = RunConfig(**run_config)
# build ofa model
ofa_model = OFA(model, run_config=run_config)
ofa_model.set_epoch(0)
ofa_model.set_task(task)
input_spec = [{
"image": paddle.ones(
shape=[1, 3, 640, 640], dtype='float32'),
"im_shape": paddle.full(
[1, 2], 640, dtype='float32'),
"scale_factor": paddle.ones(
shape=[1, 2], dtype='float32')
}]
ofa_model._clear_search_space(input_spec=input_spec)
ofa_model._build_ss = True
check_ss = ofa_model._sample_config('expand_ratio', phase=None)
# tokenize the search space
ofa_model.tokenize()
# check token map, search cands and search space
logger.info('Token map is {}'.format(ofa_model.token_map))
logger.info('Search candidates is {}'.format(ofa_model.search_cands))
logger.info('The length of search_space is {}, search_space is {}'.
format(len(ofa_model._ofa_layers), ofa_model._ofa_layers))
# set model state dict into ofa model
utils.set_state_dict(ofa_model.model, param_state_dict)
return ofa_model