Mathias Hauser

Shapely 2.0.0a is out :-) now waiting for pygeos and geopandas compatibility. I do have the feeling that there is a performance regression, though. The `STRTree` in shapely 2.0 seems...

```python import regionmask import shapely import pooch import geopandas as gp import pygeos import numpy as np file = pooch.retrieve( "https://pubs.usgs.gov/of/2006/1187/basemaps/continents/continents.zip", None ) continents = gp.read_file("zip://" + file) lon =...

There is now a pre-release of shapely v2. A compatible pygeos is releasedbut geopandas and cartopy are not yet ready. I wanted to wait for this but I am not...

I did some more test and I now think that - STRtree in shapely2 is minimally slower than in pygeos (between 0% and 10%) - STRtree in shapely2 can be...

Some more tests (after shapely/shapely#1548), running `pytest regionmask/tests/test_mask_defined_regions.py`: ``` # attrgetter (i.e. loading only): 4.0s # pygeos STRTree: 7.0s # rasterize: 10.0s # shapely 1.8 vectorized.contains: 14.3s # shapely 2.0...

~https://github.com/makepath/xarray-spatial?~

- python-rasterstats - rioxarray - clip

- xclim

yet another: https://github.com/corteva/geocube

And another [xesmf](https://pangeo-xesmf.readthedocs.io/en/latest/notebooks/Spatial_Averaging.html) and [clisops](https://clisops.readthedocs.io/en/latest/)