Robin Cole

I agree that using the callbacks will be the most intuitive approach. Do we think it's best to utilise them or just have a 'best practice' guide perhaps?

Been reading on Rastervision, perhaps we can utilise its plotting functionality with lightning callbacks - a simple tutorial could suffice https://docs.rastervision.io/en/stable/usage/tutorials/visualize_data_samples.html

Nothing wrong - I thought at some point there was a comment or issue about externalising the plotting

I'm all for tutorials - good to hear you are also mlflow user as that is also in my stack and I've made a bunch of customisations to support it,...

@DimitrisMantas good call on my ignoring the `ignore_index`.! In fairness they do address issues, but have a long backlog. When I made some noise they addressed https://github.com/Lightning-AI/torchmetrics/pull/2198 My opinion is...

Applied `on_epoch=True,` to all steps for consistency - this results in both per epoch and per step being reported **for train only** - perhaps this is why @isaaccorley did not...

@DimitrisMantas now just performing `on_step` for train loss, so a more manageable 36 metrics now

Not sure about this failing test `ValueError: Problem with given class_path 'torchgeo.trainers.SemanticSegmentationTask'`

@adamjstewart I ran the test on my branch and they pass - are they somehow run differently on CICD? ``` tests/trainers/test_segmentation.py ........................................... [100%] ================================================================================================ warnings summary ================================================================================================= tests/trainers/test_segmentation.py::TestSemanticSegmentationTask::test_trainer[True-spacenet1] tests/trainers/test_segmentation.py::TestSemanticSegmentationTask::test_trainer[True-spacenet1] /home/zeus/miniconda3/envs/cloudspace/lib/python3.10/site-packages/torch/utils/data/dataset.py:449:...

@adamjstewart tests passing