List your company
ResearchFormerly WizWand

Image Classification on ImageNet V2 (val)

80.9Top-1 AccuracyLiT-22B
10.12828.501546.87565.2485Apr 3, 2022Jul 6, 2022Oct 8, 2022Jan 10, 2023Apr 14, 2023Jul 17, 2023Oct 20, 2023
Updated 8mo ago

Evaluation results

MethodLinks
LiT-22BEvaluation Protocol=Zero-shot, Backbone=ViT-22B/142023.0280.9
CoCaEvaluation Protocol=Zero-shot2023.0280.7
BASICEvaluation Protocol=Zero-shot2023.0280.6
LiT-e/14Evaluation Protocol=Zero-shot, Backbone=ViT-e/142023.0280.6
LiT-g/14Evaluation Protocol=Zero-shot, Backbone=ViT-g/142023.0279.8
VOLO-D5 + HATParams=295.5M, FLOPs=412G, Test Size=5122022.0478.7
VOLO-D5 + HATParams=295.5M, FLOPs=304G, Test Size=4482022.0478.6
VOLO-D5Params=295.5M, FLOPs=412G, Test Size=5122022.0478
VOLO-D5Params=295.5M, FLOPs=304G, Test Size=4482022.0477.8
VOLO-D5 + HATParams=295.5M, FLOPs=69.0G, Test Size=2242022.0476.8
VOLO-D1 + HATParams=26.6M, FLOPs=22.8G, Test Size=3842022.0476.6
VOLO-D5Params=295.5M, FLOPs=69.0G, Test Size=2242022.0476.3
VOLO-D1Params=26.6M, FLOPs=22.8G, Test Size=3842022.0475.6
VOLO-D1 + HATParams=26.6M, FLOPs=6.8G, Test Size=2242022.0474.9
VOLO-D1Params=26.6M, FLOPs=6.8G, Test Size=2242022.0474
Swin-B + HATParams=87.8M, FLOPs=15.4G, Test Size=2242022.0473.8
Swin-BParams=87.8M, FLOPs=15.4G, Test Size=2242022.0472.9
Swin-S + HATParams=49.6M, FLOPs=8.7G, Test Size=2242022.0472.8
ViT-B + HATParams=86.6M, FLOPs=17.6G, Test Size=2242022.0472.6
Swin-SParams=49.6M, FLOPs=8.7G, Test Size=2242022.0472.4
Swin-T + HATParams=28.3M, FLOPs=4.5G, Test Size=2242022.0471.5
EVA CLIP-gprecision=float16, image size=224x224, zero-shot evaluation protocol=True2022.1171.5
ViT-BParams=86.6M, FLOPs=17.6G, Test Size=2242022.0471
Open CLIP-Hprecision=bfloat16, image size=224x224, zero-shot evaluation protocol=True2022.1170.9
Swin-TParams=28.3M, FLOPs=4.5G, Test Size=2242022.0470.5
ALIGNprecision=bfloat16, image size=289x289, zero-shot evaluation protocol=True2022.1170.1
CLIPEvaluation Protocol=Zero-shot2023.0270.1
ALIGNEvaluation Protocol=Zero-shot2023.0270.1
ViT-S + HATParams=22.1M, FLOPs=4.7G, Test Size=2242022.0470
OpenAI CLIP-Lprecision=float16, image size=224x224, zero-shot evaluation protocol=True2022.1169.9
Open CLIP-gprecision=bfloat16, image size=224x224, zero-shot evaluation protocol=True2022.1169.6
ViT-SParams=22.1M, FLOPs=4.7G, Test Size=2242022.0468.2
PaLI-17BShot Setting=0-shot2022.0964.46
PaLI-15BShot Setting=0-shot2022.0962.81
PaLI-3BShot Setting=0-shot2022.0962.55
ViT-T + HATParams=5.7M, FLOPs=1.6G, Test Size=2242022.0461
ViT-TParams=5.7M, FLOPs=1.6G, Test Size=2242022.0460.1
CyCLIPsmode=zero-shot, sentence_embedding=supervised2023.1018.68
CyCLIPmode=zero-shot2023.1017.77
CyCLIPnmode=zero-shot, sentence_embedding=unsupervised2023.1015.65
CLIPsmode=zero-shot, sentence_embedding=supervised2023.1014.09
CLIPnmode=zero-shot, sentence_embedding=unsupervised2023.1013.41
CLIPmode=zero-shot2023.1012.85