Image Classification on SUN397 split 1 (test)
65.6AccuracyReLICv2
Evaluation results
| Method | Links | |
|---|---|---|
| ReLICv2Backbone=ResNet-50, Pre-training dataset=JFT, Pre-training epochs=5000 ImageNet-equivalent, Evaluation protocol=Linear transfer2022.01 | 65.6 | |
| DnCBackbone=ResNet-50, Pre-training dataset=JFT, Pre-training epochs=4500 ImageNet-equivalent, Evaluation protocol=Linear transfer2022.01 | 65 | |
| ReLICv2Backbone=ResNet-50, Pre-training dataset=JFT, Pre-training epochs=1000 ImageNet-equivalent, Evaluation protocol=Linear transfer2022.01 | 64.5 | |
| BYOLBackbone=ResNet-50, Pre-training dataset=JFT, Pre-training epochs=5000 ImageNet-equivalent, Evaluation protocol=Linear transfer2022.01 | 61.8 |