Uploaded image for project: 'Data Management'
  1. Data Management
  2. DM-13671

Add option to skip cholmod.update step

    XMLWordPrintable

    Details

      Description

      The photometry model may be nonlinear enough that we cannot trivially update the Hessian after outlier rejection, but we may have to recompute it each step. This should be relatively fast for photometry. We can add an option that skips the chol.update() step in FitterBase::minimize(), and make that configurable at the python level, to test whether this improves the fitting stability.

        Attachments

          Issue Links

          is triggering

          Bug - A problem which impairs or prevents the functions of the product. DM-14552 centos6 4sigma outlier test numeric difference

          • Done

          Story - Created by Jira Software - do not edit or delete. Issue type for a user story. DM-14439 Explore numeric difference in Mac/Linux ConstrainedPhotometry fitting

          • Invalid
          relates to

          Improvement - An improvement or enhancement to an existing feature or task. DM-13669 Track and log measurement/reference outliers separately

          • Done

            Activity

            Ascending order - Click to sort in descending order
            Hide
            Permalink
            Parejkoj John Parejko added a comment -

            This should be pretty straightforward, and might help the HSC PDR1 photometry fitting failures.

            Show
            Parejkoj John Parejko added a comment - This should be pretty straightforward, and might help the HSC PDR1 photometry fitting failures.
            Hide
            Permalink
            Parejkoj John Parejko added a comment - - edited

            Jim Bosch: do you mind reviewing this (+130/-40)? It doesn't solve the problem we set out to solve, but it includes some useful code cleanups, the added astrometry test demonstrates that the implementation is sound, and we may want it for testing other things in the future.

            Jenkins run: https://ci.lsst.codes/blue/organizations/jenkins/stack-os-matrix/detail/stack-os-matrix/27965/pipeline

            Show
            Parejkoj John Parejko added a comment - - edited Jim Bosch : do you mind reviewing this (+130/-40)? It doesn't solve the problem we set out to solve, but it includes some useful code cleanups, the added astrometry test demonstrates that the implementation is sound, and we may want it for testing other things in the future. Jenkins run: https://ci.lsst.codes/blue/organizations/jenkins/stack-os-matrix/detail/stack-os-matrix/27965/pipeline
            Hide
            Permalink
            Parejkoj John Parejko added a comment -

            Thanks for the quick review. Per your comment, I added config.outlierRejectSigma, which I've been wanting for a while.

            Merged and done.

            Show
            Parejkoj John Parejko added a comment - Thanks for the quick review. Per your comment, I added config.outlierRejectSigma , which I've been wanting for a while. Merged and done.

              People

              Assignee:
              Parejkoj John Parejko
              Reporter:
              Parejkoj John Parejko
              Reviewers:
              Jim Bosch
              Watchers:
              Dominique Boutigny, Jim Bosch, John Parejko, John Swinbank, Pierre Astier
              Votes:
              0 Vote for this issue
              Watchers:
              5 Start watching this issue

                Dates

                Created:
                Updated:
                Resolved:

                  Jenkins

                  No builds found.