List your company
ResearchFormerly WizWand

Image Classification on ImageNet-1k (Custom Metrics)

88.5IN-1k AccSWAG
58.96466.63274.381.968Dec 9, 2022
Updated 8mo ago

Evaluation results

MethodLinks
SWAGArchitecture=ViT-H/14, (Pre)training data=IG-3.6B, Evaluation protocol=Fine-tuning2022.1288.59
SWAGBackbone=ViT-H/14, Pre-training Data=IG-3.6B, Evaluation Protocol=Fine-tuning2022.1288.5—
SWAGBackbone=ViT-L/16, Pre-training Data=IG-3.6B, Evaluation Protocol=Fine-tuning2022.1288—
MAEArchitecture=ViT-H/14, (Pre)training data=IN-1k, Evaluation protocol=Fine-tuning2022.1286.98
MAEBackbone=ViT-H/14, Pre-training Data=IN-1k, Evaluation Protocol=Fine-tuning2022.1286.9—
SWAGBackbone=RG-32gf, Pre-training Data=IG-3.6B, Evaluation Protocol=Fine-tuning2022.1286.8—
MAEBackbone=ViT-L/16, Pre-training Data=IN-1k, Evaluation Protocol=Fine-tuning2022.1285.9—
SWAGArchitecture=ViT-H/14, (Pre)training data=IG-3.6B, Evaluation protocol=Linear Probing2022.1285.79
SWAGBackbone=ViT-H/14, Pre-training Data=IG-3.6B, Evaluation Protocol=Linear Probing2022.1285.7—
SWAGBackbone=ViT-B/16, Pre-training Data=IG-3.6B, Evaluation Protocol=Fine-tuning2022.1285.2—
SWAGBackbone=ViT-L/16, Pre-training Data=IG-3.6B, Evaluation Protocol=Linear Probing2022.1285.1—
SWAGBackbone=RG-32gf, Pre-training Data=IG-3.6B, Evaluation Protocol=Linear Probing2022.1284.6—
MAEBackbone=ViT-B/16, Pre-training Data=IN-1k, Evaluation Protocol=Fine-tuning2022.1283.7—
SEERArchitecture=RG-32gf, (Pre)training data=IG-1B, Evaluation protocol=Fine-tuning2022.1283.39
SEERBackbone=RG-32gf, Pre-training Data=IG-1B, Evaluation Protocol=Fine-tuning2022.1283.3—
BiT-MBackbone=ResNet-50v2, Pre-training Data=IN-21k2022.1282.3—
SWAGBackbone=ViT-B/16, Pre-training Data=IG-3.6B, Evaluation Protocol=Linear Probing2022.1281.8—
Greedy SoupArchitecture=ViT-B/32, (Pre)training data=WIT, Evaluation protocol=Fine-tuning2022.12819
Greedy SoupBackbone=ViT-B/32, Pre-training Data=WIT, Evaluation Protocol=Fine-tuning2022.1281—
SupervisedBackbone=ViT-B/16, Pre-training Data=IN-1k2022.1281—
SupervisedArchitecture=RG-32gf, (Pre)training data=IN-1k2022.1280.89
SupervisedBackbone=RG-32gf, Pre-training Data=IN-1k2022.1280.8—
RobustViTBackbone=ViT-B/16, Pre-training Data=IN-1k2022.1280.3—
Uniform SoupArchitecture=ViT-B/32, (Pre)training data=WIT, Evaluation protocol=Fine-tuning2022.1279.99
Uniform SoupBackbone=ViT-B/32, Pre-training Data=WIT, Evaluation Protocol=Fine-tuning2022.1279.9—
SupervisedArchitecture=ViT-L/16, (Pre)training data=IN-1k2022.1279.68
SupervisedBackbone=ViT-L/16, Pre-training Data=IN-1k2022.1279.6—
CutMixArchitecture=ResNet-50, (Pre)training data=IN-1k2022.1278.59
CutMixBackbone=ResNet-50, Pre-training Data=IN-1k2022.1278.5—
CLIPArchitecture=ViT-H/14, (Pre)training data=LAION-2B, Evaluation protocol=zero-shot2022.1277.96
CLIPBackbone=ViT-H/14, Pre-training Data=LAION-2B, Evaluation Protocol=Zero-shot2022.1277.9—
AugMixArchitecture=ResNet-50, (Pre)training data=IN-1k2022.1277.59
AugMixBackbone=ResNet-50, Pre-training Data=IN-1k2022.1277.5—
CutoutArchitecture=ResNet-50, (Pre)training data=IN-1k2022.12778
CutoutBackbone=ResNet-50, Pre-training Data=IN-1k2022.1277—
MoCov3Backbone=ViT-B/16, Pre-training Data=IN-1k, Evaluation Protocol=Linear Probing2022.1276.6—
CLIPBackbone=ViT-G/14, Pre-training Data=LAION-2B, Evaluation Protocol=Zero-shot2022.1276.6—
CLIPArchitecture=ViT-L/14, (Pre)training data=WIT, Evaluation protocol=zero-shot2022.1276.56
CLIPBackbone=ViT-L/14, Pre-training Data=WIT, Evaluation Protocol=Zero-shot2022.1276.5—
SupervisedArchitecture=ResNet-50, (Pre)training data=IN-1k2022.1276.17
MixupArchitecture=ResNet-50, (Pre)training data=IN-1k2022.1276.17
SupervisedBackbone=ResNet-50, Pre-training Data=IN-1k2022.1276.1—
MixupBackbone=ResNet-50, Pre-training Data=IN-1k2022.1276.1—
SupervisedArchitecture=ViT-B/32, (Pre)training data=IN-1k2022.1275.99
SupervisedBackbone=ViT-B/32, Pre-training Data=IN-1k2022.1275.9—
MoCov3Architecture=ResNet-50, (Pre)training data=IN-1k, Evaluation protocol=Linear Probing2022.1274.68
MoCov3Backbone=ResNet-50, Pre-training Data=IN-1k, Evaluation Protocol=Linear Probing2022.1274.6—
CLIPArchitecture=ViT-L/14, (Pre)training data=LAION-400M, Evaluation protocol=zero-shot2022.1272.75
CLIPBackbone=ViT-L/14, Pre-training Data=LAION-400M, Evaluation Protocol=Zero-shot2022.1272.7—
Style TransferArchitecture=ResNet-50, (Pre)training data=SIN2022.1260.110
Style TransferBackbone=ResNet-50, Pre-training Data=SIN2022.1260.1—