X-Pipeline Review Meeting Monday 02 Feb 2009 09:00 Pacific / 12:00 Eastern
Agenda: Monday 02 Feb 2009 09:00 Pacific / 12:00 Eastern
- Code review: Currently at line 770 of xmakegrbwebpage.
Minutes:Attendance: Antony Searle, Xavier Siemens, Jolien Creighton, Patrick Sutton, Gareth Jones.
Code review: Currently at line 770 of xmakegrbwebpage.
- 777, 803: no discrimination on 'closedbox', 'openbox'.
- generate one random number from a large number of initial seeds and see if they are uniformly distributed.
- Not clear what is happening between 803 to 835. Goal is to have same tuning jobs as before except possibly for the selected job for ulCalc.
- Binning data, computing median in each bin, then fitting this to a quadratic, which will be smothly matched to diagonal. Why not fit to hyperbola? Why not use all data? A: safety against outliers within bins.
- reviewed xgetrunningmedian, xmedianfit, need to review xmedianbin. Got to line 969 of xmakegrbwebpage.