Dario Panici

Some more testing (all still on #1003 , without the fix Greta has, i.e. #1003 as of commit hash 533793e275a6958), running the script below gets the error ```python # running...

And like @unalmis had already mentioned, the same error occurs even if I replace the entirety of `EffectiveRipple.compute` with `return np.ones_like(self._dim_f)`

Some further sleuthing: I remember now why I pointed my finger at `rho`: if you just change the `np` calls in `EffectiveRipple` in its current state to `jnp` calls, the...

with no hidden frames: ```python env: JAX_TRACEBACK_FILTERING=off DESC version 0.12.3+1138.g533793e27.dirty,using JAX backend, jax version=0.4.31, jaxlib version=0.4.30, dtype=float64 Using device: CPU, with 3.04 GB available memory [/Users/dpanici/Research/DESC/desc/utils.py:554](http://localhost:8888/Users/dpanici/Research/DESC/desc/utils.py#line=553): UserWarning: Reducing radial (L)...

Perhaps more confusingly, if you replace ONLY `self._constants["rho"]` with a `LinearGrid(rho=rho)` (and change subsequent accesses to it with `self._constants["rho"].nodes[:,0]`), everything works fine and the error above is not received (even...

Last comment for tonight: Even if I replace every instance of `constants["rho"]` with a hard-coded `jnp.array` in the objective, if the passed-in `rho` array is placed in `constants["rho"]` in the...

This is a seemingly similar error from JAX: https://github.com/jax-ml/jax/issues/20466

Try this again without cache clearing

@gretahibbard

In our tutorial we do show doing just plt.plot: https://desc-docs.readthedocs.io/en/latest/notebooks/tutorials/EffectiveRipple.html#Calculating-effective-ripple-for-Precise-QH if right now this is not a quantity where using desc plotting is supported, we could at least add a...