DM-40136: Add AA1 metric #178
Conversation
cmsaunders
force-pushed
the
tickets/DM-40136
branch
from
76fc760
to
ce6e855
Compare
There was a problem hiding this comment.
Would it make sense to also calculate a version of AA1 that is the total positional offset? That could be useful to have sometimes.
Also, does it make sense to apply both S/N cuts (from VisitContext) and mag cuts (the “brightStar” selection)? I would think we want one or the other, but not both.
I place a few comments here, but I think we should discuss the implementation of AB1 and ABF1 on the Jira ticket. I'll post questions there.
| @@ -153,6 +153,7 @@ class FracThreshold(ScalarFromVectorAction): | |||
| threshold = Field[float](doc="Threshold to apply.") | |||
| percent = Field[bool](doc="Express result as percentage", default=False) | |||
| relative_to_median = Field[bool](doc="Calculate threshold relative to " "the median?", default=False) | |||
| absolute_value = Field[bool](doc="Take absolute value of distribution") | |||
There was a problem hiding this comment.
Should this have a default? (i.e., default=False?)
There was a problem hiding this comment.
Yes, but I'm moving it to DM-39256 anyway.
| ) | ||
|
|
||
| self.process.calculateActions.AA1 = MedianAction(vectorKey="brightStars") | ||
| self.process.calculateActions.AB1 = SigmaMadAction(vectorKey="brightStars") |
There was a problem hiding this comment.
I don't think this is what AB1 is meant to be. It should be the difference between separations measured in the r-band and those measured in any other filter. This seems to be calculating only the RMS of the separations in the input visit's filter?
There was a problem hiding this comment.
Renaming AB1 to AA1_sigmaMad.
| self.process.calculateActions.ABF1 = FracThreshold( | ||
| vectorKey="brightStars", | ||
| threshold=self.AB2, | ||
| percent=True, | ||
| op="gt", | ||
| relative_to_median=True, | ||
| absolute_value=True, | ||
| ) |
There was a problem hiding this comment.
I also think this should be calculated using separations in the visit's band relative to separations in the r-band. (i.e., it requires a pair of visits -- one in a non-r band, and one in r-band)
There was a problem hiding this comment.
Moved to DM-39256.
cmsaunders
force-pushed
the
tickets/DM-40136
branch
2 times, most recently
from
71787a1
to
1e9c3a6
Compare
| } | ||
|
|
||
| def visitContext(self) -> None: | ||
| """Apply visit options for the metrics. Applies the coadd plot flag |
There was a problem hiding this comment.
"coadd plot flag" --> "visit plot flag"?
|
|
||
|
|
||
| class TargetRefCatDeltaMetrics(AnalysisTool): | ||
| """Calculate the AA1, AB1, and ABF1 metrics from the difference between the |
There was a problem hiding this comment.
I think it's only calculating AA1 now?
cmsaunders
force-pushed
the
tickets/DM-40136
branch
from
1e9c3a6
to
f95ccbe
Compare
No description provided.