How to warm up the segmentation model upper-net?

[DZhaoXd]

Great work! I would like to train your model on my own data, but need a warm-up model for the segmentation model upper-net. Can you provide the code for training the segmentation model during the warm-up stage? Thanks

[YumengLi007]

Hi @DZhaoXd , thanks for your interests in ALDM!
For the segmenter, we used the ADE20K pretrained model from semantic-segmentation-pytorch.
Similar to the weights modification in ControlNet tool_add_control.py, you could then add one more channel in the final classification head of your trained segmenter for the extra "fake" class

[DZhaoXd]

Thank you for your positive response. I will try it.