Semi Supervised Image Classification On 2

评估指标

Top 1 Accuracy

Top 5 Accuracy

评测结果

各个模型在此基准测试上的表现结果

模型名称	Top 1 Accuracy	Top 5 Accuracy	Paper Title	Repository
WCL (ResNet-50)	72.0%	91.2%	Weakly Supervised Contrastive Learning
CoMatch + EPASS (ResNet-50)	74.1%	91.5	Debiasing, calibrating, and improving Semi-supervised Learning performance via simple Ensemble Projector
SequenceMatch (ResNet-50)	75.2%	91.9	SequenceMatch: Revisiting the design of weak-strong augmentations for Semi-supervised learning
Pseudolabeling	-	82.41%	S4L: Self-Supervised Semi-Supervised Learning
VAT + Entropy Minimization (ResNet-50)	-	83.39%	S4L: Self-Supervised Semi-Supervised Learning
PAWS (ResNet-50)	75.5%	-	Semi-Supervised Learning of Visual Features by Non-Parametrically Predicting View Assignments with Support Samples
Semi-ViT (ViT-Large)	83.3%	-	Semi-supervised Vision Transformers at Scale
VAT	-	82.78%	S4L: Self-Supervised Semi-Supervised Learning
SimCLR (ResNet-50)	-	87.8%	A Simple Framework for Contrastive Learning of Visual Representations
OBoW (ResNet-50)	-	90.7%	OBoW: Online Bag-of-Visual-Words Generation for Self-Supervised Learning
Exemplar Fine-tuned (ResNet-50)	-	81.01%	S4L: Self-Supervised Semi-Supervised Learning
Exemplar	-	81.01%	S4L: Self-Supervised Semi-Supervised Learning
SimCLRv2 (ResNet-152 x3, SK)	80.1%	95.0%	Big Self-Supervised Models are Strong Semi-Supervised Learners
Rotation Fine-tuned (ResNet-50)	-	78.53%	S4L: Self-Supervised Semi-Supervised Learning
SimCLRv2 (ResNet-50)	68.4%	89.2%	Big Self-Supervised Models are Strong Semi-Supervised Learners
Instance Discrimination	-	77.40%	Unsupervised Feature Learning via Non-Parametric Instance Discrimination	-
Rotation (joint training)	-	83.82%	Semi-supervised Sequence-to-sequence ASR using Unpaired Speech and Text	-
Meta Co-Training	85.8%	-	Meta Co-Training: Two Views are Better than One
SimMatchV2 (ResNet-50)	76.2%	-	SimMatchV2: Semi-Supervised Learning with Graph Consistency
PAWS (ResNet-50 2x)	77.8%	-	Semi-Supervised Learning of Visual Features by Non-Parametrically Predicting View Assignments with Support Samples

0 of 68 row(s) selected.