ValueError: raster.coords should be named as coordinates:(y, x) when using proximity on xarray

[kulpojke]

Hello,
I have created a binary raster from a vector file as an xarray dataset.

>>> g
<xarray.Dataset>
Dimensions:      (y: 12869, x: 6050)
Coordinates:
  * y            (y) float64 2.301e+06 2.301e+06 ... 2.173e+06 2.173e+06
  * x            (x) float64 6.315e+06 6.315e+06 ... 6.375e+06 6.375e+06
    spatial_ref  int64 0
Data variables:
    value        (y, x) float64 0.0 0.0 0.0 0.0 0.0 0.0 ... 0.0 0.0 0.0 0.0 0.0

when I try to compute proximity on g,

>>> from xrspatial import proximity
>>> prox_agg = proximity(g)

I get the following ,

---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
<ipython-input-24-d0c7776397ce> in <module>
----> 1 prox_agg = proximity(g)

/opt/conda/lib/python3.8/site-packages/xrspatial/proximity.py in proximity(raster, x, y, target_values, max_distance, distance_metric)
    771     """
    772 
--> 773     proximity_img = _process(
    774         raster,
    775         x=x,

/opt/conda/lib/python3.8/site-packages/xrspatial/proximity.py in _process(raster, x, y, target_values, max_distance, distance_metric, process_mode)
    414     raster_dims = raster.dims
    415     if raster_dims != (y, x):
--> 416         raise ValueError(
    417             "raster.coords should be named as coordinates:"
    418             "({0}, {1})".format(y, x)

ValueError: raster.coords should be named as coordinates:(y, x)

when I examine the dims

>>> g.dims
Frozen({'y': 12869, 'x': 6050})

Any ideas how I can get this to work?

[ianthomas23]

Hi @kulpojke,

xrspatial.proximity is expecting an xarray.DataArray not an xarray.Dataset. You should replace

prox_agg = proximity(g)

with

prox_agg = proximity(g.value)

This is not an uncommon problem, we should probably explicitly check that arguments are DataArrays and give a more meaningful error message if they are not.

[kulpojke]

Ok, that worked. That was so simple. Thank you.

[ianthomas23]

You're welcome!