reproduced performance is poor in swin-base

zkkli / RepQ-ViT

[ICCV 2023] RepQ-ViT: Scale Reparameterization for Post-Training Quantization of Vision Transformers

Apache License 2.0

102 stars 8 forks source link

reproduced performance is poor in swin-base #10

Open XA23i opened 2 months ago

XA23i commented 2 months ago

Hi, I try to reproduce the classification accuracy using this code. They correspond to your paper except for swin-base. I only get 68.50%, and there is a 10% gap with 78.32%. Is there any trick in swin-base model?

XA23i commented 2 months ago

does random seed get changed in swin_base model