dev(narugo): add weight_decay parameter

def train(dataset_dir: str, session_name: Optional[str] = None, from_ckpt: Optional[str] = None,

          train_ratio: float = 0.8, batch_size: int = 4, feature_bins: int = 256, fc: Optional[int] = 100,

          max_epochs: int = 500, learning_rate: LRTyping = 0.001, num_workers: Optional[int] = None,

          max_epochs: int = 500, learning_rate: LRTyping = 0.001,

          weight_decay: float = 1e-4, num_workers: Optional[int] = None,

          device: Optional[str] = None, save_per_epoch: int = 10, model_name: str = 'alexnet'):

    session_name = session_name or model_name

    _log_dir = os.path.join(_LOG_DIR, session_name)

    initial_lr = get_init_lr(learning_rate)

    optimizer = torch.optim.AdamW(

        [{'params': model.parameters(), 'initial_lr': initial_lr}],

        lr=initial_lr, weight_decay=1e-2,

        lr=initial_lr, weight_decay=weight_decay,

    )

    scheduler = get_dynamic_lr_scheduler(optimizer, lr=learning_rate, last_epoch=previous_epoch)