IPP Progress Report for the week 2012.05.14 - 2012.05.18

(Up to IPP Progress Reports)

Eugene Magnier

  • various relphot issues:
    • very bright stars (needed for ghost analysis) were failing to get reasonable magnitudes. When very saturated, these stars get broken up into multiple fainter PS1 magnitudes. My recipe for deciding to use the synthetic vs PS1 data was failing in these cases. I've made a short term fix to fall over to synthetic mags if the synthetic mags imply a very saturated star.
    • errors in the presence of ubercal data were getting screwed up. I was up-weighting ubercal values by simply dividing their errors by a factor of 10, but I failed to account for this in the weighted error calculation. I re-worked that code to have a separate weighting vector so that the error is calculated correctly.
    • there were areas of the sky which were not seeing the results of the relphot analysis. I added some per-remote client logging to track this down and found a couple of issues. First, one file on ipp008 had a hung lock, and this was crashing not just ipp008 jobs, but other ones as well (due to client / server interactions). I cleared the lock, but I'm not certain how to catch this in the future (the locking system just times out). Second, the remote clients were failing to load the image subset table because it was not yet visible over NFS. I added a flush and fsync to force the data to disk, and this helped.
  • started tests of the parallel dvomerge by running dvomerge on the WISE and SuperCOSMOS database.

Serge Chastel

Heather Flewelling

  • sick one day
  • psps
    • answered questions from nigel (not completely): there are funny problems with the stack detections
    • conrad showed me how to do the dxlayer and how to log into all the various psps machines
    • various problems with duplicate batches and other things: still sorting this out with conrad and thomas.
  • dvodbs
    • nagged people into cleaning up space on ipp005
    • stopped building dvodbs on ipp005 - will restart this week.
  • worked with JT: grab md04 smf files (specifically, the smfs corresponding to the exposures in his set of stacks) and started building a dvodb for that.

Roy Henderson

  • worked two days, up to 15th
  • PSVO:
    • new class to handle opening of streams from either online or local resource
    • trac enhancement # 370: whole plug-on menus, not just queries, can now be added dynamically
    • added 'About PSVO' text to config, added necessary code to Config class
    • added special code to make new about dialog work on Mac OS
  • ippToPsps:
    • wrote batchRepublisher.py program to republish batches that had previously failed
    • some minimal help and advice with the loading

Mark Huber

  • out sick Wednesday and Thursday
  • DRAVG minutes
  • ppStack tests - continue to clean up tests to reset for new target PSF tests. (
  • ppSub - QUB suggested additional cases, manually setting up and running test set.
  • STS refstack for MOPS - messed up initial manually selected test set, working out how to fix.

Bill Sweeney

  • Changed psphotFindDetections to replace the pass 1 sources that have been subtracted from the image after finding new peaks and before doing the footprint analysis. This greatly reduces the number of false detections near relatively bright sources.
  • Compared the results of a psphot run for a skycell with the new code with SDSS stripe 82 detections in the area. There are few objects reported by psphot that SDSS does not have a detection for.
  • Ran tests with several different recipe values for the kron measurements. Lots of mana script coding to match sources and draw plots. Plots of the results http://svn.pan-starrs.ifa.hawaii.edu/trac/ipp/wiki/Kron_measurements_investigation. Choice of whether or not to include the window function or weighting is waiting the results of testing in psphotStack (which has more sources due to forced photometery).
  • Rebuilt the mysql database used by the postage stamp sever and data store to use InnoDB tables.
  • Debugged and fixed a problem where some postage stamp requests were losing jobs. Turned out to be a bug in the sql for a revert function that was previously rarely used that now runs regularly.

Chris Waters

  • Detrend documentation wiki page: http://svn.pan-starrs.ifa.hawaii.edu/trac/ipp/wiki/GPC1_Detrend_Documentation
  • Began ROC rsync from ipp005 to ippb03.
  • Reran footprint test again.
    • Confirmed that bright stars introduce nearby false detections.
    • Checked that Bill's improvement to psphot removed these false detections.
    • Identified that remaining ghost images are due to the magnitude threshold in g not going deep enough to remove all ghosts that can form false positives.
  • AAS poster work.
  • Worked on a way to get a better initial stacking image that is faster than the KMM bimodality test, but still yields equivalent results. Need to implement and test in ppStack. I'm still somewhat concerned that the large variance due to the boost means that anything will be consistent with this initial image.