## init from some checkpoint, to resume the previous training
if args.init_from_checkpoint:
transformer.load(args.init_from_checkpoint)
## init from some pretrain models, to better solve the current task
if args.init_from_pretrain_model:
transformer.load(args.init_from_pretrain_model, reset_optimizer=True)
# model train
transformer.fit(train_data=train_loader,
eval_data=eval_loader,
epochs=args.epoch,
eval_freq=1,
save_freq=1,
save_dir=args.save_model,
callbacks=[
TrainCallback(args,
train_steps_fn=train_steps_fn,
eval_steps_fn=eval_steps_fn)
])
if __name__ == "__main__":
args = PDConfig(yaml_file="./transformer.yaml")
args.build()
args.Print()
check_gpu(args.use_cuda)
check_version()
do_train(args)
LacLoss(),
ChunkEval(num_labels),
inputs=inputs,
labels=labels,
device=args.device)
if args.init_from_checkpoint:
model.load(args.init_from_checkpoint)
if args.init_from_pretrain_model:
model.load(args.init_from_pretrain_model, reset_optimizer=True)
model.fit(train_dataset.dataloader,
epochs=args.epoch,
batch_size=args.batch_size,
eval_freq=args.eval_freq,
save_freq=args.save_freq,
save_dir=args.save_dir)
if __name__ == '__main__':
args = PDConfig(yaml_file="sequence_tagging.yaml")
args.build()
args.Print()
use_gpu = True if args.device == "gpu" else False
check_gpu(use_gpu)
check_version()
main(args)
