当前位置：网站首页>Classic backbone model cheat sheet in CV field (2)

Classic backbone model cheat sheet in CV field (2)

2022-08-07 00:24:00 【Andy Dennis】

Foreword

This article is a continuation of Cheat Sheet of Classic Backbone Models in CV Field (1).Also record some interesting backbone models.

Model

DynaMixer

ICML 2022
DynaMixer: A Vision MLP Architecture with Dynamic Mixing
Code: https://github.com/ziyuwwang/DynaMixer, 2022.8.5 Have not seen the pre-training weights announced.

block pseudocode, source original paper.

###### initializaiton #######proj_c = nn.Linear(D, D)proj_o = nn.Linear(D, D)###### code in forward ######def dyna_mixer_block(self, X):H, W, D = X.shape# row mixingfor h = 1:HY_h[h,:,:] =DynaMixerOp_h(X[h,:,:])# column mixingfor w = 1:WY_w[:,w,:] =DynaMixerOp_w(X[:,w,:])# channel mixingY_c = proj_c(X)Y_out = Y_h + Y_w + Y_creturn proj_o(Y_out)