Groupby exclude dimension

[pwolfram]

Is there some way to do a groupby operation where some dimension is excluded from the operation, e.g., a vectorized version of something like this:

vertlevels = [ds.vel.sel(nVertLevels=i).groupby('y').mean() for i in ds.nVertLevels]
xavgvel = xr.concat(vertlevels, 'nVertLevels')

The application here is to average 3D data in the x coordinate to the unique y coordinates, but not the vertical coordinate.

Thus, we are basically looking for something that allows a coordinate to be excluded from the groupby operation, e.g., in this case the vertical coordinate. Ideally this would also be possible within the context of the groupby_bins operation.

Any ideas on how this should work or a pointer on how to implement this type of operation more cleanly with existing infrastructure is greatly appreciated. This appears to be related to #324 and especially (perhaps identically) #924.

cc @vanroekel, @shoyer, @jhamman, @rabernat, @MaximilianR

[rabernat]

I explicitly listed this as a TODO in my multidimensional groupby PR (#818):
https://github.com/pydata/xarray/blob/master/xarray/core/groupby.py#L166

It would definitely be useful and is totally doable!

[shoyer]

My guess is that this could be implemented pretty quickly on top of #924, which is currently on hold while I finish up #964.

[pwolfram]

Thanks @rabernat and @shoyer!

[rabernat]

If only we could clone @shoyer!

[pwolfram]

Isn't there a git command for that? ;-)

[pwolfram]

Thanks for making everyone's life so much better via xarray and for helping us all to contribute to this process @shoyer!

[pwolfram]

Note, this issue should be resolvable via #924.

[stale]

In order to maintain a list of currently relevant issues, we mark issues as stale after a period of inactivity

If this issue remains relevant, please comment here or remove the stale label; otherwise it will be marked as closed automatically

[max-sixty]

Is this still current?

[dcherian]

Yes