Semi Supervised Image Classification On 1

评估指标

Top 1 Accuracy

评测结果

各个模型在此基准测试上的表现结果

		Paper Title
REACT (ViT-Large)	81.6%	Learning Customized Visual Models with Retrieval-Augmented Knowledge
Meta Co-Training	80.7%	Meta Co-Training: Two Views are Better than One
Semi-ViT (ViT-Huge)	80%	Semi-supervised Vision Transformers at Scale
Semi-ViT (ViT-Large)	77.3%	Semi-supervised Vision Transformers at Scale
SimCLRv2 self-distilled (ResNet-152 x3, SK)	76.6%	Big Self-Supervised Models are Strong Semi-Supervised Learners
SimCLRv2 distilled (ResNet-50 x2, SK)	75.9%	Big Self-Supervised Models are Strong Semi-Supervised Learners
MSN (ViT-B/4)	75.7%	Masked Siamese Networks for Label-Efficient Learning
SimCLRv2 (ResNet-152 x3, SK)	74.9%	Big Self-Supervised Models are Strong Semi-Supervised Learners
SimCLRv2 distilled (ResNet-50)	73.9%	Big Self-Supervised Models are Strong Semi-Supervised Learners
SimMatchV2 (ResNet-50)	71.9%	SimMatchV2: Semi-Supervised Learning with Graph Consistency
DebiasPL (ResNet-50)	71.3%	Debiased Learning from Naturally Imbalanced Pseudo-Labels
BYOL (ResNet-200 x2)	71.2%	Bootstrap your own latent: A new approach to self-supervised Learning
Semi-ViT (ViT-Base)	71%	Semi-supervised Vision Transformers at Scale
PAWS (ResNet-50 4x)	69.9%	Semi-Supervised Learning of Visual Features by Non-Parametrically Predicting View Assignments with Support Samples
PAWS (ResNet-50 2x)	69.6%	Semi-Supervised Learning of Visual Features by Non-Parametrically Predicting View Assignments with Support Samples
BYOL (ResNet-50 x4)	69.1%	Bootstrap your own latent: A new approach to self-supervised Learning
SimMatch + EPASS (ResNet-50)	68.6%	Debiasing, calibrating, and improving Semi-supervised Learning performance via simple Ensemble Projector
CoMatch + EPASS (ResNet-50)	67.4%	Debiasing, calibrating, and improving Semi-supervised Learning performance via simple Ensemble Projector
TWIST (ResNet-50 x2)	67.2%	Self-Supervised Learning by Estimating Twin Class Distributions
SimMatch (ResNet-50)	67.2%	SimMatch: Semi-supervised Learning with Similarity Matching

0 of 58 row(s) selected.