Stephan Hoyer

> To support Python 3.6 we need to sort keys for deterministic traversal order on equal but distinct dict objects, right? This is still the case even in Python 3.7+....

It looks like this version works: ``` xarray.Dataset({'x': ['foo', 'bar', 'baz']}).to_netcdf( 'foo.nc', engine='h5netcdf', encoding={'x': {'dtype': 'S1', 'zlib': True, 'chunksizes': (2, 3)}}) ``` I suppose we could update chunksizes to accept...

> In fact could the `coord_variables` dict just be an actual `xr.Coordinates` object? I think `Coordinates` should remain a thin wrapper object. It needs access to both coordinate `Variable` objects...

I am very much aligned with this revised data model! Coordinates on DataTree will still need to use a chainmap, but hopefully this is a simple variation of the base...

I'm not sure if we'd want to use exactly the same API, but an additional use-case opening a dict of Dataset objects is opening a glob/directory of netCDF files, i.e.,...

I also like open_groups! On Wed, Jul 3, 2024 at 8:04 AM Tom Nicholas ***@***.***> wrote: > I quite like open_groups! It's succinct, communicates that if you don't > have...

> Note that you can represent all data on the sphere on this grid as a square matrix On TPU/GPU, funky data access patterns can be quite slow, so I...

This is basically expected behavior. If you're supplying average forcings from an external data source like climatology or another forecast model, getting the nearest matching time would probably be the...

Hi @sit23, good to hear from you! (And sorry for the delay responding, I just saw this.) I'll see if we can work-up a demo of sharding Dinosaur across multiple...

> One of them is that the stack method applied to the velocity components here > https://github.com/google-research/dinosaur/blob/main/dinosaur/held_suarez.py#L122 > leads to trouble with the vertical padding function in spherical_harmonics So as...