Efficient Vits On Imagenet 1K With Deit T
评估指标
GFLOPs
Top 1 Accuracy
评测结果
各个模型在此基准测试上的表现结果
比较表格
| 模型名称 | GFLOPs | Top 1 Accuracy |
|---|---|---|
| spvit-enabling-faster-vision-transformers-via | 1.0 | 72.2 |
| learned-thresholds-token-merging-and-pruning | 0.8 | 71.5 |
| multi-criteria-token-fusion-with-one-step | 0.6 | 71.4 |
| patch-slimming-for-efficient-vision | 0.7 | 72.0 |
| learned-thresholds-token-merging-and-pruning | 0.7 | 69.8 |
| token-merging-your-vit-but-faster | 0.6 | 70.7 |
| patch-slimming-for-efficient-vision | 0.6 | 72.1 |
| multi-criteria-token-fusion-with-one-step | 1.0 | 72.9 |
| beyond-attentive-tokens-incorporating-token | 0.8 | 72.3 |
| token-merging-your-vit-but-faster | 0.8 | 71.4 |
| spvit-enabling-faster-vision-transformers-via | 0.9 | 72.1 |
| chasing-sparsity-in-vision-transformers-an | 0.9 | 70.1 |
| evo-vit-slow-fast-token-evolution-for-dynamic | 0.8 | 72.0 |
| scalable-visual-transformers-with | 0.6 | 69.6 |
| pruning-self-attentions-into-convolutional | 1.0 | 70.7 |
| training-data-efficient-image-transformers | 1.2 | 72.2 |
| joint-token-pruning-and-squeezing-towards | 0.8 | 72.3 |
| multi-criteria-token-fusion-with-one-step | 0.7 | 72.7 |
| learned-thresholds-token-merging-and-pruning | 1.0 | 72.0 |
| ppt-token-pruning-and-pooling-for-efficient | 0.8 | 72.1 |
| token-merging-your-vit-but-faster | 0.9 | 71.7 |
| joint-token-pruning-and-squeezing-towards | 0.8 | 72.9 |