Uploaded image for project: 'Data Management'
  1. Data Management
  2. DM-11981

Add object InputCount qa plots to CoaddAnalysis

    Details

    • Type: Story
    • Status: Done
    • Resolution: Done
    • Fix Version/s: None
    • Component/s: None
    • Labels:
      None
    • Story Points:
      8
    • Epic Link:
    • Sprint:
      DRP F18-4, DRP F18-5, DRP F18-6
    • Team:
      Data Release Production

      Description

      The base_InputCount plugin adds a base_InputCount_value field to the output catalogs that reports the doc=Number of images contributing at center, not including any clipping per object. Our HSC collaborators have noticed some potential issues with the setting of this value (namely that some objects having non-zero flux have a base_InputCount_value = 0). They have requested a plot showing the value of this field be added to the qa scripts. In particular, they would like this to be a sky view with the actual image as background if possible (i.e. if this can be made readable & informative in the context of these static plots. Some sub-selection of samples may be required to avoid overcrowding.)

        Attachments

          Activity

          Hide
          lauren Lauren MacArthur added a comment -

          This took quite a bit more fiddling than anticipated, but I think it's now good to go.  Some examples of the figures are attached and others can be seen here.

          For the issue mentioned in the description, that HSC-folk were seeing some objects having non-zero flux have a base_InputCount_value = 0, I don't see any cases in these tracts, so I artificially converted 5% of the objects to have base_InputCount_value = 0 to make sure the opaque red would show up visually. The result is:

          I've also tried to ward against having any individual source's ellipse overwhelm an area of the plot (as this would not be representative of the input count value as this only applies strictly to the centroid) by setting a maximum width or height diameter of 1000 pixels. Truncated ellipses get highlighted by being outlined with an opaque blue line. Again, there aren't any objects this large in these tracts, so I adjusted that down to one 500 pixels for one run. The result is:

          In both cases, I think the visual is striking enough to be noted by viewers of the plots (especially if they are looking for such anomalies).

          Show
          lauren Lauren MacArthur added a comment - This took quite a bit more fiddling than anticipated, but I think it's now good to go.  Some examples of the figures are attached and others can be seen  here . For the issue mentioned in the description, that HSC-folk were seeing some objects having non-zero flux have a base_InputCount_value = 0 , I don't see any cases in these tracts, so I artificially converted 5% of the objects to have base_InputCount_value = 0 to make sure the opaque red would show up visually. The result is: I've also tried to ward against having any individual source's ellipse overwhelm an area of the plot (as this would not be representative of the input count value as this only applies strictly to the centroid) by setting a maximum width or height diameter of 1000 pixels. Truncated ellipses get highlighted by being outlined with an opaque blue line. Again, there aren't any objects this large in these tracts, so I adjusted that down to one 500 pixels for one run. The result is: In both cases, I think the visual is striking enough to be noted by viewers of the plots (especially if they are looking for such anomalies).
          Hide
          lauren Lauren MacArthur added a comment -

          Would you mind giving this a look when you have a moment?

          Show
          lauren Lauren MacArthur added a comment - Would you mind giving this a look when you have a moment?
          Hide
          yusra Yusra AlSayyad added a comment -

          Thanks for turning on a linter before submitting it for review. Made it go a lot faster.

          Show
          yusra Yusra AlSayyad added a comment - Thanks for turning on a linter before submitting it for review. Made it go a lot faster.
          Hide
          lauren Lauren MacArthur added a comment -

          Thanks Yusra.  Quick update: given that I found the need for it during our investigations of the jointcal-based coadds, I added patch outlines to this plot.  Here is an example:

          Show
          lauren Lauren MacArthur added a comment - Thanks Yusra.  Quick update: given that I found the need for it during our investigations of the jointcal-based coadds, I added patch outlines to this plot.  Here is an example:

            People

            • Assignee:
              lauren Lauren MacArthur
              Reporter:
              lauren Lauren MacArthur
              Reviewers:
              Yusra AlSayyad
              Watchers:
              John Swinbank, Lauren MacArthur, Yusra AlSayyad
            • Votes:
              0 Vote for this issue
              Watchers:
              3 Start watching this issue

              Dates

              • Created:
                Updated:
                Resolved:

                Summary Panel