WP 3.5.3 H Build a prototype of parallelised post-processing pipeline for photometric redshift estimation from wide catalogues

[tms-epcc]

07/OCT/22

• waiting for LSST skymap before able to start.
• if no skymap by end of phase B then this will be deprecated.

[tms-epcc]

18/NOV/22

• Rubin DM team reported that skymap has not been made yet and that it will not happen by March'23 (ie end of Phase B).
• for WP3.5 this means the work associated with this ticket (task) will be replaced in the Oct'22 to Mar'23 plan (see action on @raphaelshirley at WP-PM meeting).

[tms-epcc]

06/JAN/23 @raphaelshirley reported

• it has been confirmed that Skymap will not be ready so title of task has changed from "Conduct preliminary image processing on LSST sky map overlap and provide detailed estimates for future run requirements" to "Build a prototype of parallelised post-processing pipeline for photometric redshift estimation from wide catalogues."
• have been experimenting with this in the last month.
• expect to complete by end of quarter
• plan to run over VIDEO

[tms-epcc]

20/JAN/23 @raphaelshirley reported

• Code now running. Now need to check against whole VIDEO field.
• Will then re-run the wide field once the catalogs exist.
• Expect computational time to be acceptable but will still require optimising to get scientifically useful results. This is expected to require additional effort in a subsequent task.
• WP3.5.3 H (ie. this task) is only concerned with developing the capability to do this.
• For code repo see https://github.com/lsst-uk/lsst-ir-fusion/tree/master/dmu6

[tms-epcc]

17/FEB/23

• prototype built and running, now checking results
• this can now be closed