Use sparse xarray for reporting.Quantity

[khaeru]

xarray 0.13 includes support for converting pd.DataFrame to a pydata/sparse data structure.
This should mostly obviate the need for the custom AttrSeries class.
A PR should be opened to make the change, test performance, and make any necessary adjustments.

Resources:

• Pint's technical commentary on container class hierarchy: https://pint.readthedocs.io/en/0.11/numpy.html#Technical-Commentary
• AmphoraInc/xarray_mongodb, which integrates xarray, sparse, and pint, for reference (thanks @gidden).

[khaeru]

#317 / iiasa/message_ix#372 partly addresses this issue; however, we still await a release of xarray >0.15.1 (see pydata/xarray#4031) to fix pydata/xarray#4019, which causes incorrect computation results in sparse-backed xarray.DataArray.

[khaeru]

we still await a release of xarray >0.15.1 (see pydata/xarray#4031) to fix pydata/xarray#4019

This is now fixed & released.

[francescolovat]

Based on your previous comment, @khaeru, can we close this issue?

[khaeru]

can we close this issue?

In short, no. That was an issue which blocked this one.

The block is now cleared, but we would still need to have a PR to implement the changes described here.

[khaeru]

This is to be continued at khaeru/genno#23.