def main():
# Initialization for the parallel enviroment
assert args.device in [
"cpu", "gpu", "xpu"
], "Invalid device! Available device should be cpu, gpu, or xpu."
paddle.set_device(args.device)
set_seed(args)
# Define the model and metric
model = BigBirdForSequenceClassification.from_pretrained(
args.model_name_or_path)
criterion = nn.CrossEntropyLoss()
metric = paddle.metric.Accuracy()
# Define the tokenizer and dataloader
tokenizer = BigBirdTokenizer.from_pretrained(args.model_name_or_path)
global config
config = getattr(model,
BigBirdForSequenceClassification.base_model_prefix).config
train_data_loader, test_data_loader = \
create_dataloader(args.batch_size, args.max_encoder_length, tokenizer)
# Define the Adam optimizer
optimizer = paddle.optimizer.Adam(parameters=model.parameters(),
learning_rate=args.learning_rate,
epsilon=1e-6)
# Finetune the classification model
do_train(model, criterion, metric, optimizer, train_data_loader, tokenizer)
# Evaluate the finetune model
do_evalute(model, criterion, metric, test_data_loader)
def main():
# Initialization for the parallel enviroment
paddle.set_device(args.device)
set_seed(args)
# Define the model and metric
# In finetune task, bigbird performs better when setting dropout to zero.
model = BigBirdForSequenceClassification.from_pretrained(
args.model_name_or_path,
attn_dropout=args.attn_dropout,
hidden_dropout_prob=args.hidden_dropout_prob)
criterion = nn.CrossEntropyLoss()
metric = paddle.metric.Accuracy()
# Define the tokenizer and dataloader
tokenizer = BigBirdTokenizer.from_pretrained(args.model_name_or_path)
config = getattr(model,
BigBirdForSequenceClassification.base_model_prefix).config
train_data_loader, test_data_loader = \
create_dataloader(args.batch_size, args.max_encoder_length, tokenizer, config)
# Define the Adam optimizer
optimizer = paddle.optimizer.Adam(parameters=model.parameters(),
learning_rate=args.learning_rate,
epsilon=1e-6)
# Finetune the classification model
do_train(model, criterion, metric, optimizer, train_data_loader, tokenizer)
# Evaluate the finetune model
do_evalute(model, criterion, metric, test_data_loader)
def setUp(self):
np.random.seed(102)
self.tokenizer = BigBirdTokenizer.from_pretrained(
'bigbird-base-uncased')
self.set_text()
self.set_input()
self.set_output()
def main():
# Initialization for the parallel enviroment
paddle.set_device(args.device)
set_seed(args)
# Define the model and metric
model = BigBirdForSequenceClassification.from_pretrained(
args.model_name_or_path)
criterion = nn.CrossEntropyLoss()
metric = paddle.metric.Accuracy()
# Define the tokenizer and dataloader
tokenizer = BigBirdTokenizer.from_pretrained(args.model_name_or_path)
global config
config = BigBirdModel.pretrained_init_configuration[
args.model_name_or_path]
train_data_loader, test_data_loader = \
create_dataloader(args.batch_size, args.max_encoder_length, tokenizer)
# Define the Adam optimizer
optimizer = paddle.optimizer.Adam(parameters=model.parameters(),
learning_rate=args.learning_rate,
epsilon=1e-6)
# Finetune the classification model
do_train(model, criterion, metric, optimizer, train_data_loader,
test_data_loader)
# Evaluate the finetune model
do_evalute(model, criterion, metric, test_data_loader)
def setUp(self):
self.tokenizer = BigBirdTokenizer.from_pretrained(
'bigbird-base-uncased')
def test_not_exist_file(self):
self.tokenizer = BigBirdTokenizer(sentencepiece_model_file='')