如何通过在python xarray中添加额外的维度将一个数组拆分为多个数组？

Question

I have the following xarray :

import xarray as xr
import numpy as np
import pandas as pd

da = xr.DataArray(np.random.rand(25, 6),
                   [('time', pd.date_range('2000-01-01', periods=25)),
                    ('space', ['IA', 'IL', 'IN', 'VA', 'VL', 'VN'])])

Where da.shape yields (25, 6) .

I would like to split the 25 into 5 x 5 such that I get a 3D shape in the following form: (5, 5, 6) .

I honestly have no idea where to start. I looked up the docs and could not find any solution. I know I seek only a vectorized solution (ie no loops and list.append() stuff, because it will be too costly otherwise.

Can you help me with a solution to this problem?

UPDATE:

With np.split(da, 5) I managed to create a list of 5 xarray s inside with exactly the data I need. How do I make this step, but instead of converting to list , it stays an xarray , but it has one extra dimension?

Answer 1

The solution I found was by using np.resize :

da_new = np.resize(da, (5,5,6))

Where (as expected) da_new.shape yields (5, 5, 6) . This works relatively quickly as well.