torch.distributed.elastic.multiprocessing.api:failed

[wxysgxxn]

When I ran it, I got this error.
I just changed the config : bash tools/dist_train.sh projects/configs/co_dino/co_dino_5scale_swin_large_16e_o365tococo.py 4 run

[wxysgxxn]

[TempleX98]

This issue open-mmlab/mmcv#1969 may help you

[zhouzhen0720]

Hello, I also encountered the same issue. Did you manage to resolve it in the end?

[wxysgxxn]

Maybe you can reduce your batch_size,when I set my batch size to half the original size, this error disappeared. I hope this can help you.

…

---Original--- From: ***@***.***> Date: Mon, Sep 18, 2023 22:09 PM To: ***@***.***>; Cc: ***@***.******@***.***>; Subject: Re: [Sense-X/Co-DETR]torch.distributed.elastic.multiprocessing.api:failed (Issue #44) Hello, I also encountered the same issue. Did you manage to resolve it in the end? — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: ***@***.***>

[zhouzhen0720]

Thank you for your reply. I have already solved this error by adding the "classes" information in the "train" section of the data dictionary in the config file.