FlexMatch: Boosting Semi-Supervised Learning with Curriculum Pseudo Labeling

Benchmark Model Rank Results
semi-supervised-image-classification-on-cifar-10-250-labelsFlexMatch#4Percentage error: 4.8±0.06
semi-supervised-image-classification-on-cifar-10-40-labelsFlexMatch#5Percentage error: 4.99±0.16
semi-supervised-image-classification-on-cifar-10-4000-labelsFlexMatch#7Percentage error: 4.19±0.01
semi-supervised-image-classification-on-cifar-100-10000-labelsFlexMatch#7Percentage error: 21.90±0.15
semi-supervised-image-classification-on-cifar-100-2500-labelsFlexMatch#7Percentage error: 26.49±0.20
semi-supervised-image-classification-on-cifar-100-400-labelsFlexMatch#10Percentage error: 39.94±1.62
semi-supervised-image-classification-on-imagenet-10-labeled-dataFlexMatch#40Top 1 Accuracy: 64.79%Top 5 Accuracy: 86.04%