Cause of warning

[Marston]

When using this example below, I keep getting this warning:

# First create an xgcm grid object
grid = Grid(ds, coords={'Z': {'center':'z'}}, periodic=False)

# define the target values in density, linearly spaced
theta_target = np.linspace(0, 3, 20)

# and transform
phi_transformed = grid.transform(ds.phi, 'Z', theta_target, target_data=ds.theta)
phi_transformed

Warning:

FutureWarning: From version 0.8.0 the Axis computation methods will be removed, in favour of using the Grid computation methods instead. i.e. use `Grid.transform` instead of `Axis.transform`
  warnings.warn(

What can I do to fix this?

[jbusecke]

You could install the latest master (we have completed the deprecation cycle and the warning is removed, but this is not released yet) or you could temporarily suppress this warning.

[jbusecke]

To be completely clear. I think this is actually a bug you discovered! Since you used the grid.transform method this warning should have never been raised. But I think this is actually fixed with the latest master.

[Marston]

That is good. I like finding and squashing bugs :-)
I already suppressed the warning.
I wonder if I could ask another question:
In the transformation code, if I understand it correctly, the grid is defined with the Z as the coordinate to transform and it seems the dataset, ds, must contain both dataArrays to be operated on. And the Z in the transform code serves the role to tell xgcm what axis to operate on, but the new da replaces the Z name with the name of the target data. Is there a way to control this? For my input da, Z name was `lev' and I would like to keep that name.

[jbusecke]

The simplest way to handle this is to rename the dimension afterwards with xarrays .rename({'oldname', 'newname'}).