How is the segmentation mask given to the model for multi label segmentation mask

[nobleaustine]

If the segmentation mask has just two class background and cancer we can give it as mask with 1 and 0s If there are multiple classes, how are they given, we can't do diffusion directly on the segmentation mask right, is onehot encoding done or any other strategy

[nobleaustine]

res = torch.where(mask > 0, 1, 0) can we run without this code in the training_losses_segmentation

[nobleaustine]

For dataset like BTCV with multiple labels how is the training done

[taolishu]

I also have the same question, can anyone answer how the mask should be entered when I do a multi-label split

[taolishu]

对于像 BTCV 这样具有多个标签的数据集，训练是如何完成的

Pls have you solved this issue at the moment