Rory Conlin

it looks like there are older versions of diffrax that work with older versions of jax. They might not have the discrete terminating event stuff but might still be useful...

@dpanici I know you can't formally approve this but lmk if you have any comments

We could probably make a simple kind of hacky thing like ```python class FixNearAxis(): def __init__(self, eq, naeq, ...): def build(self): cons = get_near_axis_constraints(self.eq, self.naeq) for con in cons: con.build()...

Something like this might be easier: ```python import inspect class Base(): """Base docstring""" def __init__(self, *, base_arg1="auto", base_arg2=None): self.base_arg1 = base_arg1 self.base_arg2 = base_arg2 # https://docs.python.org/3/reference/datamodel.html#object.__init_subclass__ def __init_subclass__(cls, /, **kwargs):...

[janssen2007zernikefft.pdf](https://github.com/PlasmaControl/DESC/files/14515753/janssen2007zernikefft.pdf)

For reference, here's a breakdown of all the unit tests and how long they take https://gist.github.com/f0uriest/dc9f2b7633dae5b84bd0d5d69e74a2f5 total time is over 2hrs

We don't do that many actual examples, I think its just DSHAPE, HELIOTRON and ATF. I don't think we can get rid of any of those without missing some important...

I think something like this is what you want: ```python import desc from desc.objectives import ObjectiveFunction, RotationalTransform, ForceBalance from desc.optimize import ProximalProjection eq = desc.examples.get("DSHAPE_CURRENT") objective = ObjectiveFunction(RotationalTransform(eq, target=1.2, loss_function="mean"))...

@abaillod did that work? Is there anything else we can help with?

Another option is given in Attenberger, Houlberg, Hirshman 1987: https://www.osti.gov/servlets/purl/6566896 Basically make R{1,0} and Z{1,0} scale like (a + offset)/a and keep the others fixed, I think for strongly shaped...