DoranLyong

I want to share what I understand about this. You can refer to Table 7 in appendix A.3 in the [paper](https://github.com/sail-sg/iFormer).  For instance with iFormer-S, each stage stacks iFormer...

The result plots are saved in the log directory.

As I can't use Baidu account, I downloaded the pretrained model from [here](https://github.com/lulujianjie/person-reid-tiny-baseline/issues/13#issuecomment-591337137). Then, I got the same problem. I found out that the pretrained model file has different layer...

@cir7 Thanks! I finally got my code :) But, I met another issue with the same of @wangyihanwyh 

I think that `self.base_model(img)` should return a tensor in (batch, dim, 7, 7) shape as input for `self.cls_head(out)`. Am I right?