My program always stops with no error, just being at a standstill, after working some epochs. And the same program works well in another dataset. By the way, GPU just run this only program. Does anybody know why? Could you give me some advice? I really appreciate that.
Is your script not responding or does it crash with some error message?
Which OS are you using?
Are you using multi-processing, e.g. for data loading (num_workers>=1)?
Is the error reproducible or does it occur randomly?
It hasn’t any error message. Just like being suspended at random epochs.
But it works well in another dataset.
I don’t use any multi-processing, and my OS is Ubuntu 16.04.