RuntimeError: copy_if failed to synchronize: device-side assert triggered

[Crystal950504]

File "train_refinedet.py", line 289, in train() File "train_refinedet.py", line 203, in train odm_loss_l, odm_loss_c = odm_criterion(out, targets) File "/home/shaoyidu/Program/anaconda3/envs/pytorch/lib/python3.6/site-packages/torch/nn/modules/module.py", line 491, in call result = self.forward(*input, **kwargs) File "/home/shaoyidu/Program/LH/RefineDet.PyTorch/layers/modules/refinedet_multibox_loss.py", line 127, in forward loss_c[pos.view(-1,1)] = 0 # filter out pos boxes for now RuntimeError: copy_if failed to synchronize: device-side assert triggered

[huangshenneng]

Excuse me, have you solved this problem?

[zeroorhero]

Excuse me, have you solved this problem? hello，have you solved the problem?Thank you!

[mhyz1n4]

Excuse me, have you solved this problem? hello，have you solved the problem?Thank you!

I've got this issue because the number of classes is defined incorrectly. Make sure to change number of calsses variables and categories variable if you are using a custom dataset. Adding CUDA_LAUNCH_BLOCKING=1 command also helps during debugging.

[whsun21]

dont have this error when using cuda 10.1, but cuda10.0 will trigger this error