RuntimeError: CUDA error: out of memory CUDA kernel errors might be asynchronously reported at some other API call,so the stacktrace below might be incorrect. For debugging consider passing CUDA_LAUNCH_BLOCKING=1.

taohan10200 / IIM

PyTorch implementations of the paper: "Learning Independent Instance Maps for Crowd Localization"

MIT License

163 stars 39 forks source link

RuntimeError: CUDA error: out of memory CUDA kernel errors might be asynchronously reported at some other API call,so the stacktrace below might be incorrect. For debugging consider passing CUDA_LAUNCH_BLOCKING=1. #28

Closed vtmjapandev closed 2 years ago

vtmjapandev commented 2 years ago

Ho guys,

How to fix this issue

RuntimeError: CUDA error: out of memory CUDA kernel errors might be asynchronously reported at some other API call,so the stacktrace below might be incorrect. For debugging consider passing CUDA_LAUNCH_BLOCKING=1.

taohan10200 commented 2 years ago

Please make sure your GPU has enough memory if your use the default experiment setting.

Tips: The training process takes ~50 hours on NWPU datasets with two TITAN RTX (48GB Memory).

If not, you would lower the batch size or use a smaller resolution.