search

transientskp / tkp

A transients-discovery pipeline for astronomical image-based surveys
http://docs.transientskp.org/
BSD 2-Clause "Simplified" License
19 stars 15 forks source link

Fixed beamwidths_limit #563

Closed ajstewart closed 5 years ago

ajstewart commented 5 years ago

The parameter beamwidths_limit is now read in from job_params.cfg and used.

Tested with my mentioned offset image in issue #562 and now associations are made as expected.

Fixes issue #562.

mkuiack commented 5 years ago

I implemented this change in my installed version 3.1.1. I reprocessed an image dataset with the same job parameters. The change didn't produce any difference in the resulting source databases. Trap is still not correctly associating nearby events

Prior to change: id | wm_ra | wm_decl | datapoints ------+------------------+------------------+------------ 5656 | 149.0172609406 | 69.5703905351843 | 261 4036 | 149.034158786887 | 69.5732569649718 | 289

After change: id | wm_ra | wm_decl | datapoints ------+------------------+------------------+------------ 5656 | 149.0172609406 | 69.5703905351843 | 261 4036 | 149.034158786887 | 69.5732569649718 | 289

Has this fixed the problem? I also notice the word "beam" doesn't appear in the association details documentation. https://tkp.readthedocs.io/en/latest/devref/database/assoc.html#database-assoc-details furthermore in the database schema under temp temprunningcatalog "beam_semimaj, beam_semimin, beam_pa Not currently used."

ajstewart commented 5 years ago

Perhaps someone can comment on the re-processing aspect, but when I ran a new job through the results were certainly different from when this fix wasn't in place. I wasn't able to associate sources with slight offsets. But adding this fix and changing the parameter to 1.5 allowed the association to happen.

As I said this was all running a new job each time when testing, and not setting the id to the existing id.