Table of Contents
Name | Direction | Type | Default | Description |
---|---|---|---|---|
InputWorkspace | Input | MatrixWorkspace | Mandatory | The input workspace. |
OutputWorkspace | Output | MatrixWorkspace | Mandatory | The result of regrouping. |
Params | Input | dbl list | Mandatory | The new approximate bin boundaries in the form: x1,dx1,x2,dx2,...,xn |
Regroups data with new bin boundaries to ensure that bins have minimum width determined by the parameter , but ensuring the bin boundaries are always coincident with original bin boundaries. The ‘params’ property defines new boundaries in intervals . Positive define constant minimum bin width, whilst negative ones create logarithmic binning
The difference between Rebin v1 and Regroup is that in the former the data in the original bins may be divided by the new bin boundaries. In the latter case, new bins are created only by combining whole bins. This is true also for the ends of the regrouped array: if the bin boundaries are 990,1010,1030,1050,...,1210, then “params” = “1000,25,1200” yields a workspace with bin boundaries 1010,1050,1090,1130,1170.
import numpy as np
# Create a workspace
ws = CreateSampleWorkspace()
# Regroup the bins such that no bin has width smaller than 300
rgws = Regroup(ws, [0,300,20000])
# Check the result
print 'Bin width in ws is',ws.readX(0)[1]-ws.readX(0)[0]
print 'Bin width in rgws is',rgws.readX(0)[1]-rgws.readX(0)[0]
# Using numpy array calculations check that all bins in the regrouped workspace
# are wider than 300
# Get the x vector of the first spectrum
x = rgws.readX(0)
# Construct an array containing the differences of the consecutive elements in array x
# x[1:] is a slice of x having all elements of x excluding the first one
# x[:-1] is a slice of x having all elements of x excluding the last one
widths = x[1:] - x[:-1]
# Check that all elements in widths are greater than 300
print np.all( widths >= 300 )
Bin width in ws is 200.0
Bin width in rgws is 400.0
True
Categories: Algorithms | Transforms | Rebin