Table of Contents
Name | Direction | Type | Default | Description |
---|---|---|---|---|
InputWorkspace | Input | MatrixWorkspace | Mandatory | The name of the input workspace. |
OutputWorkspace | Output | MatrixWorkspace | Mandatory | The name to use for the output workspace. |
WorkspaceIndex | Input | number | 0 | The index of the spectrum for fitting. |
NCoeff | Input | number | 10 | The number of b-spline coefficients. |
SplineBackground uses GSL b-spline and fitting functions to fit a spectrum. Masked bins are excluded from the fit making it possible to fit only the background signal. The output workspace has one spectrum of calculated values and the fitting errors.
import numpy as np
# Create a workspace with some data
# Fill array x with 100 float values equally spaced in the interval [-10, 10]
x = np.linspace(-10,10,100)
# Fill array y with values of a function for each number in array x. y has the same length as x
y = 1.0 / (x**2 + 10)
# Create array e of the same length as x and fill it with ones (1.0)
e = np.ones_like(x)
# Uses the above arrays to create a workspace
ws = CreateWorkspace( x, y, e )
# Run algorithm with low number of spline coefficients
out = SplineBackground(ws, WorkspaceIndex=0, NCoeff=4)
yout = out.readY(0)
# Sum the squares of the differences of elements of arrays y and yout
print 'Fit quality is',np.sum( (y - yout)**2 )
# Increase the number of spline coefficients
out = SplineBackground(ws, WorkspaceIndex=0, NCoeff=20)
yout = out.readY(0)
# Sum the squares of the differences of elements of arrays y and yout
print 'Fit quality is',np.sum( (y - yout)**2 )
Fit quality is 0.0253975992834
Fit quality is 1.53267607188e-07
Categories: Algorithms | Optimization | CorrectionFunctions\BackgroundCorrections