An interface between yt and xarray
yt_xarray streamlines communication between xarray and yt, making it easier use yt's visualization and analysis methods with data loaded via xarray.
Presently, yt_xarray primarily adds methods to generate a yt dataset from a subset of xarray dataset fields, without making copies of data when possible.
For a more detailed description, check out the full documentation at https://yt-xarray.readthedocs.io.
The latest stable version can be installed using pip with:
$ pip install yt_xarray
This will install both xarray and yt if you are missing one or the other.
The main access point is the YtAccessor object for xarray. To
use it, simply import yt_xarray and the .yt object will be available to
xarray datasets. For example, to convert the xarray dataset into a full-fledged
yt dataset:
import xarray as xr
import yt_xarray
ds = xr.open_dataset(...)
yt_ds = ds.yt.load_grid()In this example, ds.yt.grid() returns a yt dataset using all of the fields in
the xarray dataset, ready to use with any yt command. This will, however, only
work if all of your data is defined on the same grid using coordinate names that
yt understands. So for more complex cases, see the example notebooks
and the FAQ on how
to handle those cases.
Check out the example notebooks and the FAQ for examples and descriptions of common issues.
Bug reports and questions are welcome via github issues. You can also reach out via the yt slack channel (see here for how to join) by messaging Chris Havlin directly or posting in help (though you should tag @Chris Havlin in your post to get a faster response.)
Contributions are welcome, see CONTRIBUTING.md for details.