Showing content from https://github.com/xarray-contrib/datatree/issues/100 below:
Example datatree for use in tutorial documentation · Issue #100 · xarray-contrib/datatree · GitHub
What would help me enormously with writing documentation would be a killer example datatree, which I could open and use to demonstrate use of all types of methods. Just like we have the "air_temperature" example dataset used in the main xarray documentation.
To be as useful as possible, this example tree should hit a few criteria:
- Nested - there needs to be some reason why you wouldn't just use a Dataset to organise this data. Multiple resolutions is a simple reason, but it also should be >1 level deep.
- Common coordinates - it should have a least one common coordinate stored closer to the root of the tree. For example a reference normalisation value of some quantity, or perhaps some grid-related information that applies to the data in multiple sub-groups.
- Heterogenous data - there is no restriction on the relationship between data in different nodes, so we should demonstrate this by storing data that is as different as possible (but still somehow related). I'm thinking maybe some demographic data vs geographical, or model data vs observational.
- Small - though we would download this with pooch instead of uploading the data files in the repo, we still want this to be small enough that we don't cause problems when building or viewing our docs.
- Multidimensional - the data stored in the leaves needs to have enough dimensions so that I can reduce/aggregate it and still have something interesting left to plot.
- Recognisable - Ideally it would contain some relatable data. The existing Dataset example is nice because you can immediately see you are looking at a (low-resolution) map of North America. Maybe a satellite image of Manhattan Island or something?
A really good inspiration is this pseudo-structure provided in pydata/xarray#4118:
This would hit all of the criteria above, if it actually existed somewhere I could find!
What I would like is for people who have more familiarity with real geo-science data products to help me make this killer example tree, or at least point me towards data that I might use.
If we have multiple good suggestions I could make multiple different examples to use, but I think I would prefer one really good one to multiple quite good ones. Alternatively any extras could end up getting used for some future example notebooks though.
@jhamman @jbusecke @rabernat @alexamici @joshmoore
RetroSearch is an open source project built by @garambo
| Open a GitHub Issue
Search and Browse the WWW like it's 1997 | Search results from DuckDuckGo
HTML:
3.2
| Encoding:
UTF-8
| Version:
0.7.4
