Finetune Semantic Segmentation with custom model

[sirtris]

Hi,
I have pre-trained my own network backbone using the ImageEmbedder. I would now use this model in a down-stream segmentation task. How could I do this?

For a classification task I would try:
model = ImageClassifier.load_from_checkpoint("image_classification_model.pt")
The problem is that the SemanticSegmentation() has no function load_from_checkpoint

Thanks

[ethanwharris]

Hi @sirtris Thanks for the question! This is a workflow we don't currently support at the moment but is something we would like to support in the future. I'll transfer this discussion to an issue so you can track it 😃