search

NVIDIA / earth2mip

Earth-2 Model Intercomparison Project (MIP) is a python framework that enables climate researchers and scientists to inter-compare AI models for weather and climate.
https://nvidia.github.io/earth2mip/
Apache License 2.0
183 stars 40 forks source link

📚[DOC]: Information on Local Data for inference ensembles #194

Closed david5010 closed 1 month ago

david5010 commented 1 month ago

How would you describe the priority of this documentation request

Medium

Is this for new documentation, or an update to existing docs?

New

Describe the incorrect/future/missing documentation

The documentation lacks information for setting up the initial conditions. For GFS, IFS and CDS, it's rather clear. However, I'm having a lot of trouble understanding the hrmip and hdf5 files. I'm not sure how to use them, but I want to use my local data for inference_ensembles. Any help would be much appreciated!

nbren12 commented 1 month ago

The hdf5 and hrmip files are somewhat specific to our internal datasets.

If you have your own local data format, I would recommend writing your own subclass of API: https://github.com/NVIDIA/earth2mip/blob/86b11fe4ba2f19641802112e8b0ba6b962123130/earth2mip/initial_conditions/base.py#L26.

You can then use earth2mip as a library with your own datasource.

Does that make sense?

david5010 commented 1 month ago

Ohh I see, thank you!