Uploaded image for project: 'Data Management'
  1. Data Management
  2. DM-7130

Use covariance propagation code to study covariance properties

    Details

    • Type: Story
    • Status: Invalid
    • Resolution: Done
    • Fix Version/s: None
    • Component/s: None
    • Labels:
      None
    • Story Points:
      10
    • Sprint:
      DRP S17-1
    • Team:
      Data Release Production

      Description

      This means using the code developed in DM-7129 to study the results achieved on some real data. Likely start by using ci_hsc as a test dataset, and find other data as & when necessary.

      End goal is a set scatter plots showing covariance as a function of distance and a parametric fit to that. The easy case is for the "middle" of a coadd, ie with a constant number of inputs: harder, but still necessary, is to see how boundaries where the number of inputs change affect the output.

        Attachments

          Activity

          Hide
          jbosch Jim Bosch added a comment -

          I suspect we don't want to start with ci_hsc here - or at least we don't want to start with ci_hsc under its default skymap. I imagine we'll want much smaller than usual patches (and hence more of them) to be able to work with per-pixel covariance without consuming too much memory.

          Show
          jbosch Jim Bosch added a comment - I suspect we don't want to start with ci_hsc here - or at least we don't want to start with ci_hsc under its default skymap. I imagine we'll want much smaller than usual patches (and hence more of them) to be able to work with per-pixel covariance without consuming too much memory.
          Hide
          jbosch Jim Bosch added a comment -

          I also think the output needs to be a bit more than a scatter plot; I think we also want plots that show the spatial behavior of the covariance matrix, especially across discontinuities - something like an image where each pixel is some summary statistic of the row/column of the covariance matrix that corresponds to that pixel in the real image.

          Show
          jbosch Jim Bosch added a comment - I also think the output needs to be a bit more than a scatter plot; I think we also want plots that show the spatial behavior of the covariance matrix, especially across discontinuities - something like an image where each pixel is some summary statistic of the row/column of the covariance matrix that corresponds to that pixel in the real image.
          Hide
          swinbank John Swinbank added a comment - - edited

          Due to Vishal Kasliwal [X]'s likely schedule, we've agreed to put this ticket on hold and focus instead on understanding José's S/G separation code (DM-9001, DM-9002).

          Show
          swinbank John Swinbank added a comment - - edited Due to Vishal Kasliwal [X] 's likely schedule, we've agreed to put this ticket on hold and focus instead on understanding José's S/G separation code ( DM-9001 , DM-9002 ).
          Hide
          yusra Yusra AlSayyad added a comment -

          Superseded by DM-12732

          Show
          yusra Yusra AlSayyad added a comment - Superseded by  DM-12732

            People

            • Assignee:
              vpk24 Vishal Kasliwal [X] (Inactive)
              Reporter:
              swinbank John Swinbank
              Watchers:
              Jim Bosch, John Swinbank, Yusra AlSayyad
            • Votes:
              0 Vote for this issue
              Watchers:
              3 Start watching this issue

              Dates

              • Created:
                Updated:
                Resolved:

                Summary Panel