Ticket #274 (closed: fixed)

Opened 12 years ago

Last modified 5 years ago

Peak finding algorithm

Reported by: Laurent Chapon Owned by: Russell Taylor
Priority: major Milestone: Iteration 15
Component: Keywords:
Cc: Blocked By:
Blocking: Tester:

Description

Algortihm to find peaks in a 1D spectra

Change History

comment:1 Changed 12 years ago by Laurent Chapon

(In [1535]) re #274 Include <stdlib.h> and <string.h> headers to certain files. In gcc 4.3, headers dependency has been cleaned for faster compilation. Some headers have now to be explicitely included.

comment:2 Changed 12 years ago by Nick Draper

  • Owner changed from Laurent Chapon to Russell Taylor
  • Milestone changed from Iteration 13 to Iteration 14

handed over to Russell

comment:3 Changed 12 years ago by Laurent Chapon

(In [1712]) re #274 Modify error propagation in divide algorithm in the case of the error for the nominator being zero.

comment:4 Changed 12 years ago by Laurent Chapon

(In [1713]) re #274 Corrected error in calculating the wavelength from bin boundaries in the case of distributions (was x1+x2/2.0 instead of (x1+x2)/2.0.

comment:5 Changed 12 years ago by Laurent Chapon

(In [1714]) re #274 Some cleanup

comment:6 Changed 12 years ago by Laurent Chapon

(In [1715]) re #274 Some optimization for NormaliseByCurrent in the case that the input and output workspace are the same

comment:7 Changed 12 years ago by Laurent Chapon

(In [1716]) re #274 Fixed a few problems

comment:8 Changed 12 years ago by Russell Taylor

(In [1717]) Fix failing NormaliseByCurrent test. Also add FileValidators to Load/SaveNexusProcessed. Re #274.

comment:9 Changed 12 years ago by Russell Taylor

(In [1718]) Fix Load/SaveNexus test failures. Re #274.

comment:10 Changed 12 years ago by Laurent Chapon

(In [1720]) re #274 Check whether not using iteratort on workspace speeds-up computation for NormaliseByCurrent

comment:11 Changed 12 years ago by Russell Taylor

(In [1722]) Fix DivideTest. Re #274.

comment:12 Changed 12 years ago by Laurent Chapon

(In [1772]) re#274 Added an algorithm to compute the generalised second difference of spectra. This will be used by the peak-finding algorithm.

comment:13 Changed 12 years ago by Russell Taylor

(In [1773]) Add first version of FindPeaks algorithm. For now, just logs candidate peaks and doesn't perform any fits yet. Still needs tuning (i.e. debugging) because it doesn't find all peaks. Re #274.

comment:14 Changed 12 years ago by Russell Taylor

(In [1814]) (Almost) fully working version of FindPeaks algorithm. Because I can't yet pass out the list of found peaks as an output property in a TableWorkspace, I've added the stripping of peaks into this algorithm temporarily. StripPeaks should not be used at present. Re #274.

comment:15 Changed 12 years ago by Russell Taylor

(In [1820]) Include new algorithms. Re #274.

comment:16 Changed 12 years ago by Nick Draper

  • Milestone changed from Iteration 14 to Iteration 15

comment:17 Changed 12 years ago by Russell Taylor

(In [1854]) StripPeaks now uses FindPeaks as a childAlgorithm instead of using hard-coded vanadium peak positions. Re #274.

comment:18 Changed 12 years ago by Nick Draper

  • Status changed from new to closed
  • Resolution set to fixed

comment:19 Changed 5 years ago by Stuart Campbell

This ticket has been transferred to github issue ​1122

Note: See TracTickets for help on using tickets.