IPP Software Navigation Tools IPP Links Communication Pan-STARRS Links
wiki:ppSub_testing_201201

Version 10 (modified by Mark Huber, 14 years ago) ( diff )

--

Summary of ppSub_testing_201201

  1. Summary of ppSub_testing_201201
    1. Background
  2. Convolution direction choice
    1. Selected Test Samples
      1. Others Samples
  3. Monitoring
    1. Nightly tests of convolution direction
    2. HOTPANTs comparisons

Primary purpose is to wrap up and activate the convolution direction choice in ppSub.

Secondary purpose is to monitor if working as expected and address any additional issues (convolution troubles) and baseline to verify against any future changes in code (test set).

Background

Samples of previous information/work/comparisons

Convolution direction choice

We want to avoid large deconvolution in PSF-matching image subtraction. As the image quality of the images continues to improve, always convolving the reference stack will not produce the best difference image and convolving the input image is desired. The quality of the difference image is evaluated in both directions and the better is chosen for the output (add specific details).

Current branch for testing the choice of convolution direction and is updated to r33098 (from 1/11/2012)

Selected Test Samples

Initial test for if choosing the proper direction in stack-stack diffims now using MD04. Was initially done using MD10, but we want to continue to build up test samples over other IPP codes using MD04.

Start with just 2 skycells having example events from last year for photometry comparisons as well -

055 --> 077.V2 -- center -

058 --> 073.V2 -- edge - qso and A star from last year

QUB cross-match page: ​http://star.pst.qub.ac.uk/ps1/psdb/crossmatchipptocfa/

Sample for both skycells with same obsdate to look at variation of PSF center and edge.

obsdate stack_id skycell_id filter FWHM_MAJ

Difficult to use past examples of wrong convolution direction since refstacks remade, so select sample to just test current choices.

QUB exmples of wrong convolution choice:

Not all seem to be due to convolving in the wrong direction and need to identify and separate the cases/causes if can.

CfA photpipe/HOTPANTs direction: using single template/reference stack so locally get keyword CONVOL00 from header except slight issue with amps. Problem has been seems all good nightly stack samples still choose a convolved the template rather than image so nothing to test. the refstacks are very good

Sample diffims with direction forced and detections marked

event (auto direction) TARGET2 (convolved ref) TARGET1 (convolved stk) photpipe automatic
X (TARGET2) No image "test.png" attached to ppSub_testing_201201 No image "test.png" attached to ppSub_testing_201201 No image "test.png" attached to ppSub_testing_201201

Photometry comparison:

Others Samples

Need tests to verify changes will not break diffims for other surveys (MOPS etc) but do not need to worry about magic now. Not for testing if they will choose the right direction (only if using AUTO), but changes involved part of the normalization and will need to verify doesn't change flux/variance values:

  • simulated image set
  • warp-warp diffim - MD and 3PI TTI
  • warp-stack diffim - MD and LAP stack

Monitoring

Nightly tests of convolution direction

Is there any conflict with the expected desired direction? Compare to CfA photpipe, detection counts.

  • having IQ/FWHM keywords in stack images or at least the gpc1 DB would be helpful here since are calc for the PSF matching in the diffim

HOTPANTs comparisons

Difficult to automate, but are the diffims continuing to produce similar results.

  • for select night samples can setup and run locally

Attachments (26)

Note: See TracWiki for help on using the wiki.