SPViT (0.9G)

Reported on 2 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 Vision2 results

Image ClassificationonImageNet-1K (with DeiT-T)
GFLOPs· 2021-12-27
0.9
best: 1.2 (Base (DeiT-T))
SPViT: Enabling Faster Vision Transformers via Soft Token Pruning arXiv:2112.13890
Image ClassificationonImageNet-1K (with DeiT-T)
Top 1 Accuracy· 2021-12-27
72.1
best: 72.9 (dTPS)
SPViT: Enabling Faster Vision Transformers via Soft Token Pruning arXiv:2112.13890