Image Classification on CIFAR-100 Long-tailed Imbalance factor 0.01 (test)
49.73Top-1 AccResLT (3 experts)
Evaluation results
| Method | Links | |
|---|---|---|
| ResLT (3 experts)One-stage=false, Backbone=ResNet-32, experts=32021.01 | 49.73 | |
| RIDE (3 experts)One-stage=false, Backbone=ResNet-32, experts=32021.01 | 48.6 | |
| ResLTSOne-stage=true, Backbone=ResNet-32, Strong data augmentation=true2021.01 | 48.21 | |
| ResLTOne-stage=true, Backbone=ResNet-322021.01 | 45.34 | |
| Causal NormOne-stage=true, Backbone=ResNet-322021.01 | 44.1 | |
| ELF(LDAM)+DRWOne-stage=false, Backbone=ResNet-322021.01 | 43.1 | |
| BBNOne-stage=true, Backbone=ResNet-322021.01 | 42.56 | |
| LDAM+DRWOne-stage=false, Backbone=ResNet-322021.01 | 42.04 | |
| CE+DRSOne-stage=false, Backbone=ResNet-322021.01 | 41.61 | |
| CE+DRWOne-stage=false, Backbone=ResNet-322021.01 | 41.51 | |
| LDAMOne-stage=true, Backbone=ResNet-322021.01 | 39.6 | |
| CB-FocalOne-stage=true, Backbone=ResNet-32, Strong data augmentation=true2021.01 | 39.6 | |
| MixupOne-stage=true, Backbone=ResNet-32, Strong data augmentation=true2021.01 | 39.54 | |
| FocalOne-stage=true, Backbone=ResNet-32, Strong data augmentation=true2021.01 | 38.41 |