I suspect the issue stems from this https://github.com/open-mmlab/mmcv/blob/2f023453d6fc419e6ed3a8720fcf601d3863b42b/mmcv/runner/checkpoint.py#L703-L705 . Does ClearML expect the 2nd argument to torch.save to be a filename? In this case it is a BytesIO object instead
Hi NonchalantDeer14 ,
Can you share the env you are running with?
TimelyPenguin76 , env info can be found in the logs. thanks!
NonchalantDeer14 thanks for the logs, do you maybe have some toy example I can run to reproduce this issue my side?
NonchalantDeer14 which python version are you using?
sorry, i'll try to give you a toy example when i have the time to