Image Classification on ImageNet original (val) (Inductive/Transductive Accuracy)
82.05Inductive Top-1 AccuracyTransBoost
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| TransBoostArchitecture=ConvNext-T, Params (M)=28.59, Pre-training Procedure=Advanced procedure2022.05 | 82.05 | 82.46 | 0.41 | |
| TransBoostArchitecture=ViT-S, Params (M)=22.05, Pre-training Procedure=Advanced procedure2022.05 | 81.39 | 83.67 | 2.28 | |
| TransBoostArchitecture=Swin-T, Params (M)=28.29, Pre-training Procedure=Advanced procedure2022.05 | 81.38 | 82.16 | 0.78 | |
| TransBoostArchitecture=ResNet50-StrikesBack, Params (M)=25.56, Pre-training Procedure=Advanced procedure2022.05 | 80.38 | 81.15 | 0.77 | |
| TransBoostArchitecture=ResNet152, Params (M)=60.19, Pre-training Procedure=Standard simple procedure2022.05 | 78.33 | 80.64 | 2.31 | |
| SimCLRArchitecture=ResNet152, Params (M)=60.19, Learning Mode=Semi-supervised learning2022.05 | 78.33 | 78 | -0.33 | |
| SimCLRv2Architecture=ResNet152, Params (M)=60.19, Learning Mode=Semi-supervised learning2022.05 | 78.33 | 76.64 | -1.69 | |
| TransBoostArchitecture=EfficientNetB0, Params (M)=05.29, Pre-training Procedure=Advanced procedure2022.05 | 77.7 | 78.6 | 0.9 | |
| TransBoostArchitecture=ResNet101, Params (M)=44.55, Pre-training Procedure=Standard simple procedure2022.05 | 77.37 | 79.86 | 2.49 | |
| TransBoostArchitecture=ResNet50, Params (M)=25.56, Pre-training Procedure=Standard simple procedure2022.05 | 76.15 | 79.03 | 2.88 | |
| SimCLRArchitecture=ResNet50, Params (M)=25.56, Learning Mode=Semi-supervised learning2022.05 | 76.15 | 75.95 | -0.2 | |
| SimCLRv2Architecture=ResNet50, Params (M)=25.56, Learning Mode=Semi-supervised learning2022.05 | 76.15 | 74.68 | -1.47 | |
| Ent-minArchitecture=ResNet50, Params (M)=25.56, Learning Mode=Transductive few-shot learning2022.05 | 76.15 | 77.55 | 1.4 | |
| TIM-GDArchitecture=ResNet50, Params (M)=25.56, Learning Mode=Transductive few-shot learning2022.05 | 76.15 | 71.51 | -4.64 | |
| TransBoostArchitecture=MobileNetV3-L, Params (M)=05.48, Pre-training Procedure=Advanced procedure2022.05 | 75.78 | 76.81 | 1.03 | |
| TransBoostArchitecture=ResNet34, Params (M)=21.80, Pre-training Procedure=Standard simple procedure2022.05 | 73.29 | 76.7 | 3.41 | |
| TransBoostArchitecture=ResNet18, Params (M)=11.69, Pre-training Procedure=Standard simple procedure2022.05 | 69.76 | 73.36 | 3.6 |