id,node_id,number,title,user,state,locked,assignee,milestone,comments,created_at,updated_at,closed_at,author_association,active_lock_reason,draft,pull_request,body,reactions,performed_via_github_app,state_reason,repo,type
370183554,MDU6SXNzdWUzNzAxODM1NTQ=,2488,gridding data with groupby_bins in 2 dim,16838898,closed,0,,,5,2018-10-15T14:13:39Z,2020-10-04T16:01:53Z,2020-10-04T16:01:53Z,NONE,,,,"Dear everybody,

I am just starting to get to know the xarray datastructures and python so I am really still a beginner. 
I am working with scattered data wich need to be brought to a regular grid.
Now i found your function groupby_bins which only works in one dimension - on github
I couldn't find anything to wether grouping in 2d is now possible or not.
It would be very helpful to get some more info about that.

Here is just a code example with a small data set:

geop1 = 
<xarray.Dataset>
Dimensions:    (pos_compl: 44229)
Coordinates:
    lon        (pos_compl) float64 -29.8 -31.14 -32.65 ... -25.26 -16.4 -43.75
    lat        (pos_compl) float64 46.48 46.07 45.66 46.18 ... 45.34 61.06 53.19
    z          (pos_compl) float64 -3.205e+03 -3.197e+03 ... -3.758e+03
    time       (pos_compl) float64 7.299e+05 7.299e+05 ... 7.367e+05 7.367e+05
  * pos_compl  (pos_compl) complex128 (-29.805+46.485j) ... (-43.75400000000002+53.188j)
Data variables:
    geopot     (pos_compl) float64 9.363 7.93 8.218 8.621 ... 10.44 4.293 0.4243

---- groupby bins
---- 0.25
lat_bin = np.arange(mat4['lat_range'][0,0]-0.25/2,mat4['lat_range'][0,1]+0.25,0.25)
----0.5
lon_bin = np.arange(mat4['lon_range'][0,0]-0.5/2,mat4['lon_range'][0,1]+0.5,0.5)

----define bin center
---- 0.25 
lat_cent = np.arange(mat4['lat_range'][0,0],mat4['lat_range'][0,1]+0.25,0.25)
-0.5
lon_cent = np.arange(mat4['lon_range'][0,0],mat4['lon_range'][0,1]+0.5,0.5)

---- Now only these two options are possible
geop_mean_lon = geop1.geopot.groupby_bins('lon', lon_bin, labels=lon_cent)
geop_mean_lat = geop1.geopot.groupby_bins('lat', lat_bin, labels=lat_cent)

It would be really nice to have all the information in each grid box - Or is there some other way gridding like this on big datasets is recommended?

Thank you for your help!

","{""url"": ""https://api.github.com/repos/pydata/xarray/issues/2488/reactions"", ""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,completed,13221727,issue