Segmentation with Run Length encoding

Does the fastai library support segmentation with Run Length encoding? If so, could someone provide a simple example on how to do so?

Also, how would I generate predictions in run length encoding format? ?

How can this be used with SegmentationLabelList?

How did you go with this? I currently have a trained model and am attempting batch inference to generate RLEs for each class. Would be great to bounce ideas.