Ticket #3081 (closed: fixed)

Opened 9 years ago

Last modified 5 years ago

GetEi2 falls over for if the peak has small local maxima on the sides

Reported by: Martyn Gigg Owned by: Martyn Gigg
Priority: major Milestone: Iteration 29
Component: Mantid Keywords:
Cc: campbellsi@…, saviciat@… Blocked By:
Blocking: Tester: Andrei Savici

Description

If a monitor peak is sufficiently "jagged", i.e local maxima occur on the peak sides (see attached image), then this throws the GetEi algorithm off as it can't determine the peak width properly.

The attached file contains such a peak. The see the problem

  • Load the file
  • Run GetEi with monitor spectra 1 & 2 respectively and a guess of 40 MeV
  • The algorithm should fall over with a nan value for the Params of Rebin

Attachments

GetEi-Problem- Peak.png​ (21.4 KB) - added by Martyn Gigg 9 years ago.
mon_bad.nxs​ (204.9 KB) - added by Martyn Gigg 9 years ago.

Change History

Changed 9 years ago by Martyn Gigg

Changed 9 years ago by Martyn Gigg

comment:1 Changed 9 years ago by Martyn Gigg

  • Cc campbellsi@…, saviciat@… added

comment:2 Changed 9 years ago by Martyn Gigg

  • Status changed from new to accepted
  • Owner set to Martyn Gigg

comment:3 Changed 9 years ago by Martyn Gigg

  • Status changed from accepted to verify
  • Resolution set to fixed

(In [12028]) Add a check in GetEi for the bin positions it finds for the half-height of the peak. Some peaks can give rise to the algorithm giving the same bin and hence zero width. Fixes #3081

comment:4 Changed 9 years ago by Andrei Savici

  • Status changed from verify to verifying
  • Tester set to Andrei Savici

comment:5 Changed 9 years ago by Andrei Savici

  • Status changed from verifying to closed

comment:6 Changed 5 years ago by Stuart Campbell

This ticket has been transferred to github issue ​3928

Note: See TracTickets for help on using tickets.