Image Classification on Places205 (train)
56.7Top-1 AccuracySwAV
Evaluation results
| Method | Links | |
|---|---|---|
| SwAVarchitecture=R50-MLP (28M), pre-training epochs=800, evaluation augmentation=multi-crop, batch size=40962020.05 | 56.7 | |
| AMDIMarchitecture=Custom (626M), pre-training epochs=150, evaluation protocol=FastAutoAugment2020.05 | 55 | |
| PCL v2architecture=R50-MLP (28M), pre-training epochs=2002020.05 | 50.3 | |
| MoCo v2architecture=R50-MLP (28M), pre-training epochs=2002020.05 | 50.1 | |
| LocalAggarchitecture=R50 (24M), pre-training epochs=200, evaluation augmentation=10-crop2020.05 | 50.1 | |
| PIRLarchitecture=R50 (24M), pre-training epochs=8002020.05 | 49.8 | |
| PCLarchitecture=R50 (24M), pre-training epochs=2002020.05 | 49.2 | |
| MoCoarchitecture=R50 (24M), pre-training epochs=2002020.05 | 48.9 | |
| InstDiscarchitecture=R50 (24M), pre-training epochs=2002020.05 | 45.5 | |
| Rotationarchitecture=R50 (24M)2020.05 | 41.4 | |
| Jigsawarchitecture=R50 (24M), pre-training epochs=902020.05 | 41.2 | |
| DeepClusterarchitecture=VGG(15M), pre-training epochs=1002020.05 | 37.9 |