thespacedoctor
commented
7 years ago Yes - don't trust the classifications as yet, I need to tweak ranking algorithm. Currently I've hit a bottleneck with the NED downloader. I need to download NED info for ~3,000,000 transient locations before the search runs on the PS1 GW database. Sherlock is stupidly hitting NED once for each of the transient locations, which was fine for the PESSTO Marshall, but not for ATLAS and PS1. I've plan to speed this up which I'm currently implementing.
After this I will fine tune the algorithm
Star, clearly but classed as SN http://psweb.mp.qub.ac.uk/sne/ps1gw/psdb/candidate/1094132471265316100/ http://psweb.mp.qub.ac.uk/sne/ps1gw/psdb/candidate/1093142871183336500/
These should be NT, but classed as SN http://psweb.mp.qub.ac.uk/sne/ps1gw/psdb/candidate/1093641271600056100/ http://psweb.mp.qub.ac.uk/sne/ps1gw/psdb/candidate/1093117021533803600/ http://psweb.mp.qub.ac.uk/sne/ps1gw/psdb/candidate/1092159711555725700/ http://psweb.mp.qub.ac.uk/sne/ps1gw/psdb/candidate/1094819311322800600/
This is classed as NT. But the galaxy is too faint to make the judgement that the transient is "nuclear". Should be SN http://psweb.mp.qub.ac.uk/sne/ps1gw/psdb/candidate/1100840331481456600/ http://psweb.mp.qub.ac.uk/sne/ps1gw/psdb/candidate/1095252951565307300/
here's the pseudo-code I would use if host_mag < 21 then never use NT, always use SN, no matter what the separation
Should be orphan. Classed as SN. But no obvious association. http://psweb.mp.qub.ac.uk/sne/ps1gw/psdb/candidate/1093031601384323100/ http://psweb.mp.qub.ac.uk/sne/ps1gw/psdb/candidate/1094659261241219300/
These are classed as "?" But I'd prefer "orphan" if there is no contextual match http://psweb.mp.qub.ac.uk/sne/ps1gw/psdb/candidate/1082717621700310600/ http://psweb.mp.qub.ac.uk/sne/ps1gw/psdb/candidate/1072532131725709000/ http://psweb.mp.qub.ac.uk/sne/ps1gw/psdb/candidate/1065919011741247900/
Two of the last three are actually in Galaxies detected in PS1 (outside SDSS footprint) but we will need a way to ping PS1 catalogue and do star/gal separation. Not now, but will do in future)
This is classed as SN, but should be QSO http://psweb.mp.qub.ac.uk/sne/ps1gw/psdb/candidate/1100621341485244400/
Since there is an SDSS spectrum (see red box on SDSS cutout) - this comes from BOSS. Are these not in milliquas yet ? There must be a catalogue of BOSS QSOs. Since we see many.
This one doesn't have anything as a label
http://psweb.mp.qub.ac.uk/sne/ps1gw/psdb/candidate/1095614411380542300/