[🐞] <title>依次提交不同训练任务时，日志紊乱的问题。

[liuyang12a]

Is there an existing issue / discussion for this? | 是否已有关于该错误的issue或讨论？

• I have searched the existing issues / discussions | 我已经搜索过已有的issues和讨论

Is there an existing answer for this in tutorial? | 该问题是否在教程中有解答？

• I have searched tutorial | 我已经搜索过tutorial

Current Behavior | 当前行为

执行以下程序启动训练时，日志只保存在了第一个任务的检查点文件夹中：

from torch.optim.lr_scheduler import MultiStepLR

from basicts import BasicTSLauncher
from basicts.configs import BasicTSForecastingConfig
from basicts.metrics import masked_mse
from models.ASTGNN import ASTGNN, ASTGNNConfig
from basicts.runners.callback import EarlyStopping, GradientClipping
from basicts.utils import load_meta_description

def main(task_name="PEMS04"):

    data_meta = load_meta_description(task_name)

    input_len = 12
    output_len = 12


    model_config = ASTGNNConfig(
        input_len=input_len,
        output_len=output_len,
        num_features=data_meta['num_vars'],
        adj_mx_path=task_name
    )

    BasicTSLauncher.launch_training(BasicTSForecastingConfig(
        model=ASTGNN,
        model_config=model_config,
        dataset_name=task_name,
        input_len=input_len,
        output_len=output_len,
        gpus="0",
        callbacks=[EarlyStopping(), GradientClipping(1.0)], # use callbacks
        seed=233,
        num_epochs=100,
        batch_size=64,
        metrics=data_meta['regular_settings']['metrics'],
        loss=masked_mse,
        optimizer_params={
            "lr": 5e-4
        },
        lr_scheduler=MultiStepLR,
        lr_scheduler_params={
            "milestones": [25, 50],
            "gamma": 0.5
        }
    ))


if __name__ == "__main__":
    for task in ['PEMS03', 'PEMS04', 'PEMS07', 'PEMS08']:
        main(task)

Expected Behavior | 期望行为

输出的日志应当分别存在在各自训练任务的检查点文件夹中。
是否可以考虑在任务结束时增加清理日志的过程呢？

Environment | 运行环境

BasicTS logs | BasicTS日志

No response

Steps To Reproduce | 复现方法

No response

Anything else? | 备注

No response