TCDSolar / stixpy

STIX data analysis in python
https://stixpy.readthedocs.io/
BSD 3-Clause "New" or "Revised" License
17 stars 20 forks source link

Rebinnable Metadata #113

Open DanRyanIrish opened 2 months ago

DanRyanIrish commented 2 months ago

STIX observations and visibilities are associated with axis-aligned metadata that cannot be treated as a coordinate, at least not by WCS. This is because they do not represent a strictly monotonic sampling of a coordinate space. An example is detector label, "e.g. "3c, 4a,..." etc. Such metadata is not necessarily numeric and any ordering is valid.

Despite this, such metadata needs to be tracked and altered through certain operations, e.g. slicing and rebinning, in order for it to remain a faithful reflection of the data. ndcube PR #455 introduces a sliceable metadata class that performs this for slicing. However, for rebinning, axis-awareness for metadata is simply dropped as, in the generalised world of ndcube, handling this scenario is not well-defined.

This issue proposes that a STIXVisMeta that inherents from ndcube.Meta and VisMeta in xrayvision, and a STIXMeta for metadata associated with STIX observations. Extra functionality should be added to handle STIX-specific metadata during the rebinning process.

As part of this discussion, this issue should catalogue all the axis-aligned STIX metadata that is specific to observations, visibilities and both that can also be treated during rebinning, e.g. detector label.

Both (STIXMetaBase)

STIXMeta

STIXVisMeta