Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Semi-supervised Classification on ImageNet 10% labels (train)
Loading...
0.765
Top-1 Accuracy
TWIST
0.55596
0.61023
0.6645
0.71877
Oct 14, 2021
Top-1 Accuracy
Top-5 Accuracy
Updated 1mo ago
Evaluation Results
Method
Method
Links
Top-1 Accuracy
Top-5 Accuracy
TWIST
Backbone=ViT-B/16
2021.10
0.765
0.931
TWIST
Backbone=ResNet-50x2
2021.10
0.753
0.928
DINO
Backbone=ViT-B/16
2021.10
0.746
0.92
BYOL
Backbone=ResNet-50x2
2021.10
0.735
0.917
TWIST
Backbone=ResNet-50
2021.10
0.717
0.91
SimCLR
Backbone=ResNet-50x2
2021.10
0.717
0.912
SwAV
Backbone=ResNet-50
2021.10
0.702
0.899
BarlowTwins
Backbone=ResNet-50
2021.10
0.697
0.893
BYOL
Backbone=ResNet-50
2021.10
0.688
0.89
DINO
Backbone=ResNet-50
2021.10
0.682
0.891
SimCLR
Backbone=ResNet-50
2021.10
0.656
0.878
SUP
Backbone=ResNet-50
2021.10
0.564
0.804
Feedback
Search any
task
Search any
task
