List your company
ResearchFormerly WizWand

Image Classification on ImageNet ReaL 6 (val)

0.91AccuracySoViT-400m/14
0.902720.904610.90650.90839May 22, 2023
Updated 8mo ago

Evaluation results

MethodLinks
SoViT-400m/14Pretraining=JFT-3B, Input resolution=518x518, Parameters=428M, FLOPs=1374G, Fine-tuning=true2023.050.91
SoViT-400m/14Pretraining=JFT-3B, Input resolution=384x384, Parameters=428M, FLOPs=672G, Fine-tuning=true2023.050.909
ViT-g/14Pretraining=JFT-3B, Input resolution=518x518, Parameters=1011M, FLOPs=3208G, Fine-tuning=true2023.050.909
MAE-WSPPretraining=IG-3B, Input resolution=518x518, Parameters=1890M, FLOPs=5679G, Fine-tuning=true2023.050.909
ViT-G/14Pretraining=JFT-3B, Input resolution=518x518, Parameters=1882M, FLOPs=5668G, Fine-tuning=true2023.050.908
ViT-L/16Pretraining=JFT-3B, Input resolution=384x384, Parameters=303M, FLOPs=383G, Fine-tuning=true2023.050.904
SoViT-400m/14Pretraining=JFT-3B, Input resolution=224x224, Parameters=428M, FLOPs=221G, Fine-tuning=true2023.050.903