huggingface的参数使用(HfArgumentParser+parser.parse_args_into_dataclasses+TrainingArguments)

import argparse def args_parser(): parser = argparse.ArgumentParser() ### new parser.add_argument('--exp', type=str, default='VEIN/UNET_FL', help='experiment_name') parser.add_argument('--labeled_num', type=int, default=1000, help='labeled data') parser.add_argument('--model', type=str, default='unet_fl', help='model_name') parser.add_argument('--num_classes', type=int, default=2, help='output channel of network') parser.add_argument('--gpu', type=str, default='0,1', help='GPU to use') parser.add_argument('--deterministic', type=int, default=1, help='whether use deterministic training') parser.add_argument('--seed', type=int, default=1337, help='random seed') parser.add_argument('--root_path', type=str, default='/data3/vein/vein-train-1000-480-640', help='dataset root dir') parser.add_argument('--num_users', type=int, default=10, help='local epoch') parser.add_argument('--batch_size', type=int, default=4, help='batch_size per gpu') parser.add_argument('--base_lr', type=float, default=1e-4, help='segmentation network learning rate') parser.add_argument('--max_iterations', type=int, default=1000, help='maximum epoch number to train') parser.add_argument('--rounds', type=int, default=50, help='local epoch') parser.add_argument('--local_ep', type=int, default=1, help='local epoch') parser.add_argument('--dataset_seg', type=str, default='vein', help='dataset') parser.add_argument('--in_chan', type=int, default=1, help='input channel') parser.add_argument('--base_chan', type=int, default=32, help='base channel') parser.add_argument('--out_chan', type=int, default=2, help='out channel') parser.add_argument('--reduce_size', type=list, default=[9,12], help='reduce size') #parser.add_argument('--local_train', type=int, default=1000, help='local train time') parser.add_argument('--root_path_unlabeled', type=str, default='/data3/vein/vein-train-unlabel-2000-480-640', help='dataset root dir of unlabeled data') parser.add_argument('--each_length', type=int, default=225, help='data number each length') parser.add_argument('--dist_scale', type=int, default=0.01, help='scale factor when computing model distance') parser.add_argument("--max_grad_norm", dest="max_grad_norm", type=float, default=5, help="max gradient norm allowed (used for gradient clipping)") parser.add_argument('--num-warmup-epochs', '--num-warm-up-epochs', dest="num_warmup_epochs", default=0, type=int, help='number of warm-up epochs for unsupervised loss ramp-up during training' 'set to 0 to disable ramp-up') parser.add_argument('--lambda_u', type=float, default=100, help='start_epoch') parser.add_argument('--ema_decay', type=float, default=0.999, help='ema_decay') parser.add_argument('--consistency', type=float, default=0.1, help='consistency') parser.add_argument('--consistency_rampup', type=float, default=200.0, help='consistency_rampup') parser.add_argument('--labeled_bs', type=int, default=4, help='labeled_batch_size per gpu') parser.add_argument('--ict_alpha', type=int, default=0.2, help='ict_alpha') args = parser.parse_args() return args