Xarray_SimpleUnits

Unitful calculations with xarray

Keeps track of units when working with xarray.DataArrays and xarray.Datasets using astropy.units for the conversions.

:warning: This package is in alpha stage, that is, it works mostly, but the interface might still be subject to change.

Install

Requirements

• astropy - required
• xarray - required
• pytest - optional, for testing

xarray_simpleunits

An installable pip package called xarray-simpleunits will be available soon from the main package repository, it can then be installed with:

$ pip install xarray_simpleunits

The latest development version can be installed with pip directly from github (see https://pip.pypa.io/en/stable/reference/pip_install/#vcs-support and https://pip.pypa.io/en/stable/reference/pip_install/#git):

$ pip install [-e] git+https://github.com/st-bender/xarray-simpleunits.git

The other option is to use a local clone:

$ git clone https://github.com/st-bender/xarray-simpleunits.git
$ cd xarray-simpleunits

and then using pip (optionally using -e, see https://pip.pypa.io/en/stable/reference/pip_install/#install-editable):

$ pip install [-e] .

or using setup.py:

$ python setup.py install

Optionally, test the correct function of the module with

$ py.test [-v]

or even including the doctests in this document:

$ py.test [-v] --doctest-glob='*.md'

Usage

The python module itself is named xarray_simpleunits and is imported as usual.

All functions should be numpy-compatible and work with scalars and appropriately shaped arrays.

>>> import xarray_simpleunits as xru

The module basically works by adapting (“monkey-patching”) the xarray.Variable arithmetic methods to honour and keep track of the "units" attribute. To initialize unit handling with xarray, call init_units() first:

>>> import xarray_simpleunits as xru
>>> xru.init_units()

(A similar method to restore the original behaviour is planned.)

So far, the supported operations are addition, subtraction, multiplication, and division. Unit mismatch when adding or subtracting unitful arrays raises an exception. Currently, unit handling with xarray.DataArray requires that the respective array is on the left side of any calculation:

>>> from astropy import units as au
>>> import numpy as np
>>> import xarray as xr
>>> import xarray_simpleunits as xru
>>> np.set_printoptions(precision=6)
>>> xru.init_units()
>>> ds = xr.Dataset(
...     data_vars={
...         "s": ("x", [1., 2., 3.], {"units": "m"}),
...         "t": ("x", [3., 2., 1.], {"units": "s"}),
...     },
... )
>>> v = ds["s"] / ds["t"]
>>> v  # doctest: +ELLIPSIS
<xarray.DataArray (x: 3)>...
array([0.333333, 1.      , 3.      ])
Dimensions without coordinates: x
Attributes:
    units:    m / s
>>> # using `astropy` units directly:
>>> v = ds["s"] / (2 * au.Unit("s"))
>>> v  # doctest: +ELLIPSIS
<xarray.DataArray 's' (x: 3)>...
array([0.5, 1. , 1.5])
Dimensions without coordinates: x
Attributes:
    units:    m / s

Basic class and method documentation is accessible via pydoc:

$ pydoc xarray_simpleunits

License

This python package is free software: you can redistribute it or modify it under the terms of the GNU General Public License as published by the Free Software Foundation, version 2 (GPLv2), see local copy or online version.