Image Classification on ImageNet 10% labeled 1k (test)
91.2Top-5 AccuracyWCL
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| WCLLearning Paradigm=Self-Supervised, Fine-tuning Layer=Projection Head2021.10 | 91.2 | 72 | — | |
| WCLLearning Paradigm=Self-Supervised, Fine-tuning Layer=AvgPool2021.10 | 90.3 | 71.1 | — | |
| FNCancelLearning Paradigm=Self-Supervised, Fine-tuning Layer=Projection Head2021.10 | 90.2 | 71.1 | — | |
| SwAVLearning Paradigm=Self-Supervised, Fine-tuning Layer=AvgPool2021.10 | 89.9 | 70.2 | — | |
| FixMatchLearning Paradigm=Semi-supervised2021.10 | 89.1 | 71.46 | — | |
| BYOLLearning Paradigm=Self-Supervised, Fine-tuning Layer=AvgPool2021.10 | 89 | 68.8 | — | |
| UDALearning Paradigm=Semi-supervised2021.10 | 88.5 | — | — | |
| SimCLR v1Learning Paradigm=Self-Supervised, Fine-tuning Layer=AvgPool2021.10 | 87.8 | 65.6 | — | |
| PCLLearning Paradigm=Self-Supervised, Fine-tuning Layer=AvgPool2021.10 | 86.2 | — | — | |
| S4LLearning Paradigm=Semi-supervised2021.10 | 83.8 | — | — | |
| PIRLLearning Paradigm=Self-Supervised, Fine-tuning Layer=AvgPool2021.10 | 83.8 | 57.2 | — | |
| SupervisedLearning Paradigm=Supervised2021.10 | 80.4 | 48.4 | — | |
| InstDiscLearning Paradigm=Self-Supervised, Fine-tuning Layer=AvgPool2021.10 | 77.4 | — | — | |
| DeiT-STraining Setting=Elite, Epochs=3002021.04 | — | — | 80.06 | |
| DeiT-STraining Setting=Elite, Epochs=3002021.04 | — | — | 40.41 | |
| DeiT-TiTraining Setting=Elite, Epochs=3002021.04 | — | — | 78.72 | |
| DeiT-TiTraining Setting=Elite, Epochs=3002021.04 | — | — | 38.44 | |
| ResNet-50Training Setting=Elite, Epochs=3002021.04 | — | — | 89.9 | |
| ResNet-50Training Setting=Elite, Epochs=3002021.04 | — | — | 58.37 | |
| ResNet-50-55%Training Setting=Elite, Epochs=3002021.04 | — | — | 87.1 | |
| ResNet-50-55%Training Setting=Elite, Epochs=3002021.04 | — | — | 51.48 | |
| SimCLR v2Learning Paradigm=Self-Supervised, Fine-tuning Layer=Projection Head, Backbone=ResNet-502021.10 | — | 68.4 | — | |
| SimCLR v2Learning Paradigm=Self-Supervised, Fine-tuning Layer=Projection Head, Backbone=ResNet-1012021.10 | — | 71.4 | — | |
| Visformer-STraining Setting=Elite, Epochs=3002021.04 | — | — | 90.06 | |
| Visformer-STraining Setting=Elite, Epochs=3002021.04 | — | — | 58.74 | |
| Visformer-TiTraining Setting=Elite, Epochs=3002021.04 | — | — | 89.48 | |
| Visformer-TiTraining Setting=Elite, Epochs=3002021.04 | — | — | 55.14 |