How to train 3D images using the SegResNetVAE model?

[Minxiangliu]

I enjoy using models and transforms within MONAI for my medical image research. Currently, I'd like to try training 3D images using the SegResNetVAE model, but I've noticed that it has two outputs (images and loss values), so the examples in the MONAI tutorial don't seem to be applicable. Could someone here please suggest some examples for training with SegResNetVAE? Thanks in advance!