This repository was archived by the owner on Sep 11, 2023. It is now read-only.

Description
It's possible that we might want to remove datetime features from the on-disk batches, and instead compute datetime features on-the-fly, not least because we'll want to experiment with a variety of different ways of encoding position when we really start experimenting with the perceiver IO models (see openclimatefix/nowcasting_utils#30)
These datetime features should be really fast to compute. And, if we start saving different 'modalities' to disk as different files, then it will almost certainly be faster to compute these on-the-fly rather than saving them into tiny files on disk.