Amshaker / SwiftFormer

[ICCV'23] Official repository of paper SwiftFormer: Efficient Additive Attention for Transformer-based Real-time Mobile Vision Applications
247 stars 25 forks source link

Fix this bug when setting distillation-type to none #12

Closed ThomasCai closed 10 months ago

ThomasCai commented 10 months ago

Hello, after I raised this issue #11 , I tried to solve it and found that this is the problem. Please help me see it if this solution is appropriate. Thank you.