List your company
ResearchFormerly WizWand

Image Classification on ImageNet-1k 1.0 (val) (Top-1/Top-5 Accuracy, Top-1 Loss)

83Top-1 AccuracySwin-S
68.02471.91275.879.688Nov 23, 2021
Updated 8mo ago

Evaluation results

MethodLinks
Swin-SFLOPs (G)=8.7, Params (M)=49.62021.1183—96.2
SPViT-Swin-SFLOPs (G)=6.1, Params (M)=38.9, fine-tuning with knowledge distillation=true2021.1183096.4
SPViT-DeiT-BFLOPs (G)=8.4, Params (M)=41.6, fine-tuning with knowledge distillation=true2021.1182.4-0.696.1
SPViT-Swin-SFLOPs (G)=6.1, Params (M)=38.92021.1182.40.696
MDC-DeiT-BFLOPs (G)=11.22021.1182.3-0.5—
S2ViTE-DeiT-BFLOPs (G)=11.7, Params (M)=56.8, sparse training from scratch=true2021.1182.2-0.4—
EVIT-DeiT-BFLOPs (G)=11.6, Params (M)=86.42021.1182.1-0.395.6
DeiT-BFLOPs (G)=17.5, Params (M)=86.42021.1181.8—95.6
SPViT-DeiT-BFLOPs (G)=8.4, Params (M)=41.62021.1181.50.395.7
DynamicViT-DeiT-BFLOPs (G)=11.4, Params (M)=87.12021.1181.40.495.5
VTP-DeiT-BFLOPs (G)=13.8, Params (M)=67.32021.1181.30.595.3
IA-RED2-DeiT-BFLOPs (G)=11.8, Params (M)=86.42021.1181.30.5—
dTPS-DeiT-BFLOPs (G)=11.4, Params (M)=872021.1181.20.6—
Swin-TiFLOPs (G)=4.5, Params (M)=28.32021.1181.2—95.5
eTPS-DeiT-BFLOPs (G)=11.4, Params (M)=86.42021.1181.10.7—
SPViT-Swin-TiFLOPs (G)=3.4, Params (M)=25.8, fine-tuning with knowledge distillation=true2021.11810.295.4
VTP-DeiT-BFLOPs (G)=10, Params (M)=482021.1180.71.1—
UVC-DeiT-BFLOPs (G)=8, fine-tuning with knowledge distillation=true2021.1180.61.2—
SPViT-DeiT-SFLOPs (G)=3.3, Params (M)=15.9, fine-tuning with knowledge distillation=true2021.1180.3-0.495.1
dTPS-DeiT-SFLOPs (G)=3, Params (M)=22.82021.1180.1-0.2—
SSP-DeiT-BFLOPs (G)=11.7, Params (M)=56.8, sparse training from scratch=true2021.1180.11.7—
SPViT-Swin-TiFLOPs (G)=3.4, Params (M)=25.82021.1180.11.195
DeiT-SFLOPs (G)=4.6, Params (M)=22.12021.1179.9—95
MDC-DeiT-SFLOPs (G)=2.92021.1179.90—
eTPS-DeiT-SFLOPs (G)=3, Params (M)=22.12021.1179.70.2—
STEP-Swin-SFLOPs (G)=6.3, Params (M)=36.92021.1179.63.494.7
EVIT-DeiT-SFLOPs (G)=3, Params (M)=22.12021.1179.50.494.8
ToMe-DeiT-SFLOPs (G)=2.7, Params (M)=22.12021.1179.40.5—
UVC-DeiT-SFLOPs (G)=2.7, fine-tuning with knowledge distillation=true2021.1179.40.5—
DynamicViT-DeiT-SFLOPs (G)=3, Params (M)=22.82021.1179.30.694.7
S2ViTE-DeiT-SFLOPs (G)=3.2, Params (M)=14.6, sparse training from scratch=true2021.1179.20.7—
SPViT-DeiT-SFLOPs (G)=3.3, Params (M)=15.92021.1178.31.694.3
SSP-DeiT-SFLOPs (G)=3.2, Params (M)=14.6, sparse training from scratch=true2021.1177.72.2—
STEP-Swin-TiFLOPs (G)=3.5, Params (M)=23.62021.1177.2493.6
SPViT-DeiT-TiFLOPs (G)=1, Params (M)=4.8, fine-tuning with knowledge distillation=true2021.1173.2-191.4
DeiT-TiFLOPs (G)=1.3, Params (M)=5.72021.1172.2—91.1
UVC-DeiT-TiFLOPs (G)=0.7, fine-tuning with knowledge distillation=true2021.1171.80.4—
SPViT-DeiT-TiFLOPs (G)=1, Params (M)=4.82021.1170.71.590.3
S2ViTE-DeiT-TiFLOPs (G)=1, Params (M)=4.2, sparse training from scratch=true2021.1170.12.1—
SSP-DeiT-TiFLOPs (G)=1, Params (M)=4.2, sparse training from scratch=true2021.1168.63.6—