rbiswas4 / run2_diaproc

LSSTDESC Processing steps for Run2

Run2 DIA process

This repo contains several aids for script creation, that make use of the drivers available in dia_pipe repo.
The intention is to have a guide so anyone can build template images from calexps, and after that perform DIA analysis.
The scripts provided work as a cli, and they would ingest a tract and a patch numbers, and produce shell files with the commands so the user can run this at NERSC.

Several options are available, but the user needs to specify the following information:

• database: the tracts_mapping.sqlite3 SQLite database for tract/patch selection
• calexp_repo: the calexp butler repository, so we can use the images there to produce template coadds.
• tract, patch
• Depending on the step of the analysis you will need to set the templ or template repo, and the dia_repo

Steps to carry the analysis

• Create templates
  • Pick a tract-patch in Run2
  • Obtain the list of visits and filters for this patch
  • Configure the coaddDriver.py
  • Run the coaddDriver.py
  • Run the multiBandDriver.py on the coadd repo
• Run the DIA pipeline
  • Configure the chosen DIA algorithm
  • Run imageDifferenceDriver.py
  • Obtain the DIASources and DIAObjects by means of associationDriver.py
  • Filter the detections to create sensible catalogs
  • Optionally create stamps, lightcurves and ...
• Study results
  • Obtain a matching in space-time between the DIAObjects and truth catalogs
  • Obtain metric figures quantifying the completitude-contamination
  • Measure selection functions
  • Etc.