# xrspatial.classify.quantile¶

xrspatial.classify.quantile(agg: xarray.core.dataarray.DataArray, k: int = 4, name: Optional[str] = 'quantile') xarray.core.dataarray.DataArray[source]

Reclassifies data for array agg into new values based on quantile groups of equal size.

Parameters
• agg (xarray.DataArray) – 2D NumPy, CuPy, NumPy-backed Dask, or Cupy-backed Dask array of values to be reclassified.

• k (int, default=4) – Number of quantiles to be produced.

• name (str, default='quantile') – Name of the output aggregate array.

Returns

quantile_agg – 2D aggregate array, of quantile allocations. All other input attributes are preserved.

Return type

xarray.DataArray, of the same type as agg

Notes

• Dask’s percentile algorithm is approximate, while numpy’s is exact.

• This may cause some differences between results of vanilla numpy

References

Examples

```import numpy as np
import xarray as xr
from xrspatial.classify import quantile

elevation = np.array([
[np.nan,  1.,  2.,  3.,  4.],
[ 5.,  6.,  7.,  8.,  9.],
[10., 11., 12., 13., 14.],
[15., 16., 17., 18., 19.],
[20., 21., 22., 23., np.inf]
])
data = xr.DataArray(elevation, attrs={'res': (10.0, 10.0)})
data_quantile = quantile(data, k=5)
```
```>>> print(data)
<xarray.DataArray (dim_0: 5, dim_1: 5)>
array([[nan,  1.,  2.,  3.,  4.],
[ 5.,  6.,  7.,  8.,  9.],
[10., 11., 12., 13., 14.],
[15., 16., 17., 18., 19.],
[20., 21., 22., 23., inf]])
Dimensions without coordinates: dim_0, dim_1
Attributes:
res:      (10.0, 10.0)

>>> print(data_quantile)
<xarray.DataArray 'quantile' (dim_0: 5, dim_1: 5)>
array([[nan,  0.,  0.,  0.,  0.],
[ 0.,  1.,  1.,  1.,  1.],
[ 2.,  2.,  2.,  2.,  2.],
[ 3.,  3.,  3.,  3.,  4.],
[ 4.,  4.,  4.,  4., nan]], dtype=float32)
Dimensions without coordinates: dim_0, dim_1
Attributes:
res:      (10.0, 10.0)
```