Image Classification on ImageNet OOD DEEPNETS-1M (test)
92.9Top-5 Accuracy (RESNET)SGD (100 epochs)
Evaluation results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| SGD (100 epochs)Number of updates=1000k, GPU compute time (sec)=6e5*, CPU compute time (sec)=6e7*2021.10 | 92.9 | — | — | — | — | 72.2 | |
| SGD (10000 steps)Number of updates=10k, GPU compute time (sec)=4e3, CPU compute time (sec)=6e42021.10 | 49 | 38.7 | 22.1 | 36.3 | 8 | 33.4 | |
| SGD (5000 steps)Number of updates=5k, GPU compute time (sec)=2e3, CPU compute time (sec)=3e42021.10 | 34.8 | 26.2 | 13.2 | 25.4 | 4.8 | 24.3 | |
| GHN-1Number of updates=1, GPU compute time (sec)=0.3, CPU compute time (sec)=0.52021.10 | 6.9 | 15.8 | 15.9 | 15.1 | 0.5 | 0.9 | |
| GHN-2Number of updates=1, GPU compute time (sec)=0.3, CPU compute time (sec)=0.72021.10 | 5.3 | 19.4 | 24.7 | 26.4 | 7.2 | 4.4 | |
| SGD (1 step)Number of updates=1, GPU compute time (sec)=0.4, CPU compute time (sec)=6.02021.10 | 0.5 | 0.5 | 0.5 | 0.5 | 0.5 | 0.5 |