List your company
ResearchFormerly WizWand

Image Classification on ImageNet 1K (val) (FT Accuracy)

86.2FT AccuracyMIRL
82.14483.19784.2585.303Sep 25, 2023
Updated 8mo ago

Evaluation results

MethodLinks
MIRLEncoder=ViT-B-48, #params=341M, FLOPs=67.0G, Training Data=IN1K, Epochs=800, Resolution=224x2242023.0986.2
MAEEncoder=ViT-L, #params=307M, FLOPs=59.7G, Training Data=IN1K, Epochs=1600, Resolution=224x2242023.0985.9
HPMEncoder=ViT-L, #params=307M, FLOPs=59.7G, Training Data=IN1K, Epochs=800, Resolution=224x2242023.0985.8
MaskFeatEncoder=ViT-L, #params=307M, FLOPs=59.7G, Training Data=IN1K, Epochs=1600, Resolution=224x2242023.0985.7
ConMIMEncoder=ViT-L, #params=307M, FLOPs=59.7G, Training Data=IN1K, Epochs=1600, Resolution=224x2242023.0985.5
MAEEncoder=ViT-L, #params=307M, FLOPs=59.7G, Training Data=IN1K, Epochs=800, Resolution=224x2242023.0985.4
MIRLEncoder=ViT-B-48, #params=341M, FLOPs=67.0G, Training Data=IN1K, Epochs=300, Resolution=224x2242023.0985.3
MIRLEncoder=ViT-S-54, #params=96M, FLOPs=18.8G, Training Data=IN1K, Epochs=800, Resolution=224x2242023.0984.8
MIRLEncoder=ViT-B-24, #params=171M, FLOPs=33.5G, Training Data=IN1K, Epochs=300, Resolution=224x2242023.0984.7
MAEEncoder=ViT-L, #params=307M, FLOPs=59.7G, Training Data=IN1K, Epochs=300, Resolution=224x2242023.0984.5
MIRLEncoder=ViT-S-54, #params=96M, FLOPs=18.8G, Training Data=IN1K, Epochs=300, Resolution=224x2242023.0984.4
MIRLEncoder=ViT-B, #params=86M, FLOPs=16.8G, Training Data=IN1K, Epochs=800, Resolution=224x2242023.0984.1
LocalMIMEncoder=ViT-B, #params=86M, FLOPs=16.8G, Training Data=IN1K, Epochs=1600, Resolution=224x2242023.0984
SimMIMEncoder=ViT-B, #params=86M, FLOPs=16.8G, Training Data=IN1K, Epochs=800, Resolution=224x2242023.0983.8
MAEEncoder=ViT-B, #params=86M, FLOPs=16.8G, Training Data=IN1K, Epochs=1600, Resolution=224x2242023.0983.6
MIRLEncoder=ViT-B, #params=86M, FLOPs=16.8G, Training Data=IN1K, Epochs=300, Resolution=224x2242023.0983.5
CIMEncoder=ViT-B, #params=86M, FLOPs=16.8G, Training Data=DALLE250M+IN1K, Epochs=300, Resolution=224x2242023.0983.3
MoCov3Encoder=ViT-B, #params=86M, FLOPs=16.8G, Training Data=IN1K, Epochs=300, Resolution=224x2242023.0983.2
BEITEncoder=ViT-B, #params=86M, FLOPs=16.8G, Training Data=DALLE250M+IN1K, Epochs=800, Resolution=224x2242023.0983.2
MAEEncoder=ViT-B, #params=86M, FLOPs=16.8G, Training Data=IN1K, Epochs=800, Resolution=224x2242023.0983.1
MAEEncoder=ViT-B, #params=86M, FLOPs=16.8G, Training Data=IN1K, Epochs=300, Resolution=224x2242023.0982.6
SupervisedEncoder=ViT-L, #params=307M, FLOPs=59.7G, Training Data=IN1K, Epochs=N/A, Resolution=224x2242023.0982.6
SupervisedEncoder=ViT-B, #params=86M, FLOPs=16.8G, Training Data=IN1K, Epochs=N/A, Resolution=224x2242023.0982.3