DynamicViT (90%)

Reported on 4 benchmarks across 1 task · 1 paper

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Computer Vision4 results

Image ClassificationonImageNet-1K (With LV-ViT-S)
GFLOPs· 2021-06-03
5.8
best: 6.6 (Base (LV-ViT-S))
DynamicViT: Efficient Vision Transformers with Dynamic Token Sparsification arXiv:2106.02034
Image ClassificationonImageNet-1K (With LV-ViT-S)
Top 1 Accuracy· 2021-06-03
83.3
best: 83.5 (MCTF ($r=8$))
DynamicViT: Efficient Vision Transformers with Dynamic Token Sparsification arXiv:2106.02034
Image ClassificationonImageNet-1K (with DeiT-S)
GFLOPs· 2021-06-03
4
best: 4.6 (Base (DeiT-S))
DynamicViT: Efficient Vision Transformers with Dynamic Token Sparsification arXiv:2106.02034
Image ClassificationonImageNet-1K (with DeiT-S)
Top 1 Accuracy· 2021-06-03
79.8
best: 80.1 (MCTF ($r=16$))
DynamicViT: Efficient Vision Transformers with Dynamic Token Sparsification arXiv:2106.02034