CTShower Testing - Mar. 1, 2005

Mirror of the best candidate showers page.

The following are reconstructions with the code I inherited from Shawn, with no alterations. Version 2.35 is the chicos chi-squared fit, version 3.01 is the debugged version of the maximum likelihood fit 3.00 on the web, and version 3.02 is the maximum likelihood fit using the second set of probability functions. I didn't use the START/SOTC options on any of these fits (in fact it doesn't seem to work on lancer anyway, for unknown reasons), but they have accidentals and IGN sites removed. The maximum likelihood fit should be able to deal with accidentals, but it is useful for comparison purposes to take them out for now.

01-29-03 v2.35 (OK)
v3.01 (similar, but lower LDF)
v3.02 (stuck on a site)

03-03-03 v2.35 (misses)
v3.01 (OK, but low LDF)
v3.02 (similar, low LDF)

03-24-03 v2.35 (not great)
v3.01 (better)
v3.02 (fails to stop)

04-19-03 v2.35 (not great)
v3.01 (a bit better)
v3.02 (fails; returns NANs)

05-13-03 v2.35 (stuck on a site)
v3.01 (OK, but low LDF)
v3.02 (fails to stop)

05-17-03 v2.35 (stuck on a site)
v3.01 (OK, but low LDF)
v3.02 (fails to stop)

06-14-03 v2.35 (OK)
v3.01 (fails; returns NANs)
v3.02 (fails to stop)

06-18-03 v2.35 (stuck on a site)
v3.01 (offset, with low LDF)
v3.02 (fails to stop)

06-27-03 v2.35 (OK)
v3.01 (fails; returns NANs)
v3.02 (fails; returns NANs)

08-17-03 v2.35 (not great)
v3.01 (similar)
v3.02 (similar)

11-20-03 v2.35 (stuck on a site)
v3.01 (OK, but low LDF)
v3.02 (OK)

12-26-03 v2.35 (not great)
v3.01 (fails; returns NANs)
v3.02 (fails; returns NANs)