Open tensorflowt opened 2 years ago
We haven't made such a comparison.
For ImageNet classification 224x224, the computation costs of Swin and ours are not in the same order of magnitude. To some extent, it is unnecessary to compare Swin and ours.
Method | FLOPs | Params | Top-1 |
---|---|---|---|
Swin-Tiny | 4.5G | 29M | 81.3 |
Topformer-B | 0.37G | 5M | 75.3 |
Hi!Excuse me ,I have some question to ask you ?
Hi! Excuse me, have you compared the speed and accuracy of topformer and swin transformer under the same computing power? Looking forward to your reply. good luck!