*cudnn.benchmark = True
- This flag allows you to enable the inbuilt cudnn auto-tuner to find the best algorithm to use for your hardware.
- benchmark mode is good whenever your input sizes for your network do not vary. This way, cudnn will look for the optimal set of algorithms for that particular configuration (which takes some time). This usually leads to faster runtime.
반응형
'스타트업 > AI' 카테고리의 다른 글
| [AI] squeeze(), unsqueeze() (0) | 2020.02.04 |
|---|---|
| [AI] Mutil-GPU(in Pytorch) (0) | 2020.02.04 |
| [AI] pytorch save, load, state_dict (0) | 2020.02.03 |
| [AI] radam (0) | 2020.02.03 |
| [AI] 상대 경로, 절대 경로 (0) | 2020.01.31 |