how to use xarray like pandas panel when adding new items

Question

I have converted pandas panel to xarray but cannot add new items, major axis and minor axis as easily as I can with pandas panel. The code is below:

import numpy as np

import pandas as pd

import xarray as xr


panel = pd.Panel(np.random.randn(3, 4, 5), items=['one', 'two', 'three'], 
                 major_axis=pd.date_range('1/1/2000', periods=4),
                 minor_axis=['a', 'b', 'c', 'd','e'])

if I want to add a new item for example, I can:

panel.four=pd.DataFrame(np.ones((4,5)),index=pd.date_range('1/1/2000', periods=4), columns=['a', 'b', 'c', 'd','e'])

panel.four

            a   b   c   d   e
2000-01-01  1.0 1.0 1.0 1.0 1.0

2000-01-02  1.0 1.0 1.0 1.0 1.0

2000-01-03  1.0 1.0 1.0 1.0 1.0

2000-01-04  1.0 1.0 1.0 1.0 1.0

I have difficulty in increasing the items, major/minor axis in xarray

px=panel.to_xarray()

#px gives me
<xarray.DataArray (items: 3, major_axis: 5, minor_axis: 4)>

array([[[-0.440081, -0.888226,  0.158702,  2.107577],
        [ 0.917835, -0.174557,  0.501626,  0.116761],
        [ 0.406988,  1.95184 , -1.345948,  2.960774],
        [-1.905529,  0.25793 ,  0.076162,  1.954012],
        [ 0.499675,  1.87567 , -1.698771, -1.143766]],


       [[ 0.070269, -1.151737, -0.344155, -0.506383],
        [-2.199357, -0.040909,  0.491984, -0.333431],
        [-0.113155, -0.668475,  2.366683, -0.421863],
        [-0.567336, -0.302224,  1.638386, -0.038545],
        [ 0.55067 , -0.409266, -0.27916 , -0.942144]],


       [[ 1.269171, -0.151471, -0.664072,  0.269168],
        [-0.486492,  0.59632 , -0.191977,  0.22537 ],
        [ 0.069231, -0.345793, -0.450797, -2.982   ],
        [-0.42338 , -0.849736,  0.965738, -0.544596],
        [-1.455378, -0.256441, -1.204572, -0.347749]]])

Coordinates:

  * items       (items) object 'one' 'two' 'three'

  * major_axis  (major_axis) datetime64[ns] 2000-01-01 2000-01-02 2000-01-03 ...

  * minor_axis  (minor_axis) object 'a' 'b' 'c' 'd'


#how should I add a fourth item, increase/delete major axis, minor axis?

Answer 1

xarray assignments are not as elegant as the pandas panel. Lets say we want to add a fourth item in the data array above. Here is how it works:

four=xr.DataArray(np.ones((1,4,5)), coords=[['four'],pd.date_range('1/1/2000', periods=4),['a', 'b', 'c', 'd','e']], 
                  dims=['items','major_axis','minor_axis'])

pxc=xr.concat([px,four],dim='items')

Whether the operation is on items or major/minor axis, a similar logic prevails. For deleting use

pxc.drop(['four'], dim='items')