what's the difference between Attention-56 and Attention-92(MX)

[wuyuuu]

Hi,

I found the trained models in https://github.com/JDAI-CV/FaceX-Zoo/tree/main/training_mode, where you offer Attention-56 and Attention-92(MX).

However, when I looked at backbone codes backbone and backboneconf, I only saw a single model and configuration.

As I am gonna use your models as feature extraction backbones, I am wondering how to initialize those two backbones (Attention-56 and Attention-92(MX)) properly.

Looking forward to your reply.
Thanks!

[wang21jun]

They share the same backbone, but the configuration is different.
AttentionNet56:
AttentionNet:
stage1_modules: 1
stage2_modules: 1
stage3_modules: 1
feat_dim: 512
out_h: 7
out_w: 7

AttentionNet92:
AttentionNet:
stage1_modules: 1
stage2_modules: 2
stage3_modules: 3
feat_dim: 512
out_h: 7
out_w: 7

[wuyuuu]

They share the same backbone, but the configuration is different. AttentionNet56: AttentionNet: stage1_modules: 1 stage2_modules: 1 stage3_modules: 1 feat_dim: 512 out_h: 7 out_w: 7

AttentionNet92: AttentionNet: stage1_modules: 1 stage2_modules: 2 stage3_modules: 3 feat_dim: 512 out_h: 7 out_w: 7

Get it, thanks for your answer!