question about cls_modules

[freshman2001]

Hi! Thank you so much for this valuable work for the community. I would like to ask, in line 73 of clrnet/model/head/clrhead.py, is the role of self.cls_modules to judge whether the priority contains lane line samples?

[MySuperSoul]

Hi! Thank you so much for this valuable work for the community. I would like to ask, in line 73 of clrnet/model/head/clrhead.py, is the role of self.cls_modules to judge whether the priority contains lane line samples?

Not exactly, our head consists of two branchs, one branch for classification and another for regression. The cls_modules is used to generate classification-related feature from original extracted lane feature, which consists of three LinearModule. After cls_modules there will be one linear layerself.cls_layers = nn.Linear(self.fc_hidden_dim, 2) to output cls result.