DM-53441: Create an analysis_tool to extract first-order astrometry summary stats for visit images#467
DM-53441: Create an analysis_tool to extract first-order astrometry summary stats for visit images#467jrmullaney wants to merge 3 commits intomainfrom
Conversation
jrmullaney
force-pushed
the
tickets/DM-53441
branch
from
5c5dfdf to
b91f1e5
Compare
| outputUnit="arcminute", | ||
| ) | ||
| self.process.buildActions.ratio = DivideVector() | ||
| self.process.buildActions.ratio.actionA = AngularSeparation( |
There was a problem hiding this comment.
Can you use the already calculated version of this from above?
There was a problem hiding this comment.
Good catch - I'll sort that out.
There was a problem hiding this comment.
Oh - on second thoughts, I don't think I can without introducing a filterActions step, which doesn't seem right for this calculation (it's not a filter).
We can't guarantee what order the buildActions will be done in, so we can't know for certain that cornersep will be built before the DivideVector is attempted.
So I think it's better to leave as-is, but let me know otherwise.
| from ..interfaces import AnalysisBaseConfig, AnalysisBaseConnections, AnalysisPipelineTask | ||
|
|
||
|
|
||
| class SimpleTableAnalysisConnections( |
There was a problem hiding this comment.
I am still not convinced we need this and that one of the other table readers couldn't be adapted. If we do need this then the name should probably be changed before someone thinks it reads Simple AFW tables.
There was a problem hiding this comment.
There isn't a an existing table reader that reads-in an ExposureCatalog and just passes it straight to run. There are ones that read in an ExposureCatalog, but they do loads of manipulation prior to passing it to run. self.LoadData doen't work with an Exposurecatalog, so I can't use an existing ArrowAstropy reader that just passes to run.
So, I think I have to make a new ExposureCatalog reader that passes straight to run anyway.
But the justification for doing this was that thare are a fair few table readers that basically do the same job - read in an ArrowAstropy then pass it straight to run. But they are all in different files, which could all be consolidated.
If we want, I could just make this into a VisitSummaryAnalysisTask for now, but I do think we should consider whether to consolidate all the tasks that basically just read and pass-on ArrowAstropy tables.
There was a problem hiding this comment.
Good points, let's add consolidation of table tasks to the future to do list.
The atool calculates the on-sky corner-to-corner distance of each detector then reports summary stats for a whole visit. It also reports summary stats for pixelScales across the whole visit.
jrmullaney
force-pushed
the
tickets/DM-53441
branch
from
b91f1e5 to
02e36c8
Compare
2 participants
No description provided.