For metacal, we want to be able to generate realistic galaxies, and it would be good to be able to tune the morphologies from simple and smooth to pretty disturbed, as a way of spanning a range of possible shapes and check that metacal works on these. Erin's current procedure involves using parametric + knots light profiles, we could try to demonstrate that our generative models are more realistic than that using our set of morphology statistics.
For metacal, we want to be able to generate realistic galaxies, and it would be good to be able to tune the morphologies from simple and smooth to pretty disturbed, as a way of spanning a range of possible shapes and check that metacal works on these. Erin's current procedure involves using parametric + knots light profiles, we could try to demonstrate that our generative models are more realistic than that using our set of morphology statistics.