- ``--job_id``: The job unique id, it affects the log files' name. e.g., ``--job_id=job1``. Default ``--job_id=default``.
- ``--devices``: The selected accelerate devices on nodes, can be gpu/xpu/npu etc.. e.g., ``--devices=0,1,2,3`` will launch four training processes each bound to one device.
- ``--devices``: The selected accelerate devices on nodes, can be gpu/xpu etc.. e.g., ``--devices=0,1,2,3`` will launch four training processes each bound to one device.
- ``training_script``: The full path to the single GPU training program/script to be launched in parallel, followed by all the arguments for the training script. e.g., ``training.py``