Ticket #853 (closed: fixed)

Opened 11 years ago

Last modified 5 years ago

Convolution algorithm

Reported by: Nick Draper Owned by: Roman Tolchenov
Priority: major Milestone: Iteration 23
Component: Keywords:
Cc: Blocked By:
Blocking: Tester:

Description

We would like a program that convolutes/folds/blurs a specified analytic function with the resolution/blurring function and fits to the measured data. This should be similar in operation to the existing fitting routines. The resolution could be in 1 of 3 forms:

  1. An analytic function
  2. An ASCII file giving a single curve
  3. A workspace with similar structure to the measured input data

For options 1 & 2, all histograms use the same resolution function. For option 3, each input histogram has its corresponding resolution function.

Change History

comment:1 Changed 11 years ago by Nick Draper

  • Milestone changed from Iteration 19 to Iteration 20

comment:2 Changed 11 years ago by Nick Draper

  • Owner set to Roman Tolchenov

comment:3 Changed 11 years ago by Roman Tolchenov

(In [3261]) Correction to fourier transform. re #853

comment:4 Changed 11 years ago by Roman Tolchenov

(In [3264]) A correction to FFTSmooth algorithm. re #853

comment:5 Changed 11 years ago by Roman Tolchenov

(In [3267]) Another correction to RealFFT. Added RealFFT Test. re #853

comment:6 Changed 11 years ago by Roman Tolchenov

(In [3268]) Temporarily disabled FFTTest. re #853

comment:7 Changed 11 years ago by Roman Tolchenov

(In [3282]) Fix to FFTTest. re #853

comment:8 Changed 11 years ago by Roman Tolchenov

  • Status changed from new to accepted

comment:9 Changed 11 years ago by Roman Tolchenov

Speak to Felix Fernandez-Alonso to check that his convolution requirements match our understanding.

comment:10 Changed 11 years ago by Roman Tolchenov

(In [3791]) Added Convolution function. re #853

comment:11 Changed 11 years ago by Roman Tolchenov

(In [3821]) Temporarily hid Convolution. re #853

comment:12 Changed 11 years ago by Roman Tolchenov

(In [3888]) Added ConstraintFactory and modified FunctionFactory to be able to create nested composite functions. re #853

comment:13 Changed 11 years ago by Russell Taylor

(In [3889]) Fix for Linux. Re #853.

comment:14 Changed 11 years ago by Roman Tolchenov

(In [3931]) Added Resolution. re #853

comment:15 Changed 11 years ago by Roman Tolchenov

(In [3938]) Changed dll export of ParameterFactory. re #853

comment:16 Changed 11 years ago by Roman Tolchenov

(In [3939]) Forgot that linux does not use dllexport. re #853

comment:17 Changed 11 years ago by Roman Tolchenov

(In [3941]) Changes to constraints to make their use simpler. re #853

comment:18 Changed 11 years ago by Russell Taylor

(In [3944]) Replace reference to Workspace2D with one to MatrixWorkspace. Re #853.

comment:19 Changed 11 years ago by Roman Tolchenov

(In [3945]) Moved constraint application to Fit. re #853

comment:20 Changed 11 years ago by Roman Tolchenov

(In [4061]) Integrated changed ties and constraints into MnatidPlot. re #853

comment:21 Changed 11 years ago by Roman Tolchenov

(In [4144]) Made it possible to use Convolution in MantidPlot. re #853

comment:22 Changed 11 years ago by Roman Tolchenov

(In [4170]) Fixed a few bugs in the fit browser, added save and load function commands. re #853

comment:23 Changed 10 years ago by Roman Tolchenov

(In [4272]) Corrections to FunctionFactory to handle brackets better. re #853

comment:24 Changed 10 years ago by Roman Tolchenov

(In [4273]) Forgot to correct this test. Also made Fit Function property InOut. re #853

comment:25 Changed 10 years ago by Roman Tolchenov

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

Work continued in #1200

comment:26 Changed 10 years ago by Nick Draper

Anders, during testing, check what is implemented and raise seperate tickets for the rest.

comment:27 Changed 10 years ago by Anders Markvardsen

  • Status changed from testing to closed

Implemented and documented: 1. An analytic function 2. An ASCII file giving a single curve

May not be implemented: 3. A workspace with similar structure to the measured input data For options 1 & 2, all histograms use the same resolution function. For option 3, each input histogram has its corresponding resolution function.

comment:28 Changed 5 years ago by Stuart Campbell

This ticket has been transferred to github issue ​1701

Note: See TracTickets for help on using tickets.