Classification on Stanford Cars 15% labels (test)
78.76AccuracySelf-Tuning+HCR
Evaluation results
| Method | Links | |
|---|---|---|
| Self-Tuning+HCRBackbone=ResNet-50, Pre-trained=ImageNet, Label Proportion=15%2022.06 | 78.76 | |
| Self-TuningBackbone=ResNet-50, Pre-trained=ImageNet, Label Proportion=15%2022.06 | 74.99 | |
| Mean TeacherBackbone=ResNet-50, Pre-trained=ImageNet, Label Proportion=15%2022.06 | 54.28 | |
| FixMatchBackbone=ResNet-50, Pre-trained=ImageNet, Label Proportion=15%2022.06 | 49.86 | |
| SimCLRv2Backbone=ResNet-50, Pre-trained=ImageNet, Label Proportion=15%2022.06 | 45.74 | |
| Pi-modelBackbone=ResNet-50, Pre-trained=ImageNet, Label Proportion=15%2022.06 | 45.19 | |
| Pseudo-LabelingBackbone=ResNet-50, Pre-trained=ImageNet, Label Proportion=15%2022.06 | 40.93 | |
| UDABackbone=ResNet-50, Pre-trained=ImageNet, Label Proportion=15%2022.06 | 39.9 |