Error when training the model

[jwxsp1]

Hi, when I try to train the model, I have an error: Do you know how to fix it?

[ccsasuke]

Hi @jwxsp1, as the error message mentioned, you're using the slurm trainer, which will produce an error if SLURM is not found. If your cluster does not use slurm, you should change to a different trainer such as gpu_1_host.

[jwxsp1]

Hi @jwxsp1, as the error message mentioned, you're using the slurm trainer, which will produce an error if SLURM is not found. If your cluster does not use slurm, you should change to a different trainer such as gpu_1_host.

Hello, thank you for your reply. I have tried to replace the trainer, but there seems to be a new problem. How can I solve this problem? I would appreciate it if you would like to answer it.

[ccsasuke]

@jwxsp1 This line in the default config is supposed to address this error. Are you loading the configs correctly? Our project uses Hydra to manage configs and command-line arguments in case you'd like to learn more about how to change configs such as the trainer.