You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Dataset jdn022r[361elt]q. Files for debugging: jdn022r3q stored in █████████████████
16 November 2023
Differences noted between HSTDP-2023.20 and HSTDP-2023.30 where the
WCS got "worse" in the latest build.
When looking at the FLC images, The GAIA sources are well identified.
The GSC242 sources are not as distinct and there is no overlap in the
identified sources with GAIA. The GAIA results are essentially the same
between the two regression tests. However, the TOTAL RMS ('max_rms')
found for the GSC242 improved between the processings. The preferred
threshold is for the TOTAL RMS <= 20.0 which none of the processings
achieved. The GSC242 presumably improved because some suspect sources
were culled (num_matches 46 -> 32). What I am wondering about is:
(1) how the RA and DEC RMS values are computed with respect to the
TOTAL RMS as the GSC242 result got much worse, and
(2) what threshold changed to decrease the number of presumably
bogus matches in the GSC242 case.
Issue HLA-1165 was created on JIRA by Michele De La Pena:
Dataset jdn022r[361elt]q. Files for debugging: jdn022r3q stored in █████████████████
16 November 2023
Differences noted between HSTDP-2023.20 and HSTDP-2023.30 where the
WCS got "worse" in the latest build.
When looking at the FLC images, The GAIA sources are well identified.
The GSC242 sources are not as distinct and there is no overlap in the
identified sources with GAIA. The GAIA results are essentially the same
between the two regression tests. However, the TOTAL RMS ('max_rms')
found for the GSC242 improved between the processings. The preferred
threshold is for the TOTAL RMS <= 20.0 which none of the processings
achieved. The GSC242 presumably improved because some suspect sources
were culled (num_matches 46 -> 32). What I am wondering about is:
(1) how the RA and DEC RMS values are computed with respect to the
TOTAL RMS as the GSC242 result got much worse, and
(2) what threshold changed to decrease the number of presumably
bogus matches in the GSC242 case.
=====
HSTDP-2023.2.0
GAIAeDR3 relative {'valid': True, 'max_rms': 25.447, 'num_matches': 22, 'compromised': False, 'reason': '', 'fit_qual': 3}
GSC242 relative {'valid': True, 'max_rms': 28.460, 'num_matches': 46, 'compromised': False, 'reason': '', 'fit_qual': 3}
2MASS relative {'valid': True, 'max_rms': 100000, 'num_matches': 0, 'compromised': False, 'reason': '', 'fit_qual': 5}
GAIAeDR3 22.637 27.408
GSC242 29.851 41.522
=====
HSTDP-2023.3.0
GAIAeDR3 relative {'valid': True, 'max_rms': 25.499, 'num_matches': 22, 'compromised': False, 'reason': '', 'fit_qual': 3}
GSC242 relative {'valid': True, 'max_rms': 24.336, 'num_matches': 32, 'compromised': False, 'reason': '', 'fit_qual': 3}
2MASS relative {'valid': True, 'max_rms': 100000, 'num_matches': 0, 'compromised': False, 'reason': '', 'fit_qual': 5}
GAIAeDR3 22.593 27.539
GSC242 42.966 49.917
The text was updated successfully, but these errors were encountered: