MedianBinWidth dialog.
Table of Contents
| Name | Direction | Type | Default | Description |
|---|---|---|---|---|
| InputWorkspace | Input | MatrixWorkspace | Mandatory | The workspace containing the input data |
| Rounding | Input | string | None | Bin width rounding. Allowed values: [‘None’, ‘10^n’] |
| BinWidth | Output | number | The averaged median bin width |
This algorithm calculates the median bin width of each histogram in InputWorkspace. The (optionally rounded) mean value of the medians is then placed in the BinWidth output property.
If the Rounding property is set to 10^n, the bin width will be rounded down to the nearest power of 10. For example, 0.11 and 0.99 will be rounded to 0.1, while 0.011 and 0.099 will be rounded to 0.01.
The InputWorkspace has to contain histogram data. For point data, ConvertToHistogram v1 can be used first, but care should be taken if the points are not equally spaced.
Example: rebinning a workspace.
import numpy
# For normal python lists: 3 * [0.1] = [0.1, 0.1, 0.1]
binWidths = 3 * [0.1] + 4 * [1.7] + [10.0]
# Convert to numpy array.
# For numpy arrays: 3 * [0.1] = [0.3], thus the above magic would not work.
binWidhts = numpy.array(binWidths)
# Make bin boundaries out of the widths. The first boundary is at -3.0.
xs = numpy.cumsum(numpy.append(numpy.array([-3.0]), binWidths))
# There is one less bin than the number of boundaries.
ys = numpy.zeros(len(xs) - 1)
ws = CreateWorkspace(DataX=xs, DataY=ys)
newWidth = MedianBinWidth(InputWorkspace=ws)
print('New bin width: {0}'.format(newWidth))
rebinned = Rebin(InputWorkspace=ws, Params=[newWidth], FullBinsOnly=True)
widths = ws.readX(0)[1:] - ws.readX(0)[:-1]
print('Bin widths before rebinning: {0}'.format(widths))
widths = rebinned.readX(0)[1:] - rebinned.readX(0)[:-1]
print('Bin widths after rebinning: {0}'.format(widths))
Output:
New bin width: 1.7
Bin widths before rebinning: [ 0.1 0.1 0.1 1.7 1.7 1.7 1.7 10. ]
Bin widths after rebinning: [ 1.7 1.7 1.7 1.7 1.7 1.7 1.7 1.7 1.7 1.7]
Categories: AlgorithmIndex | Utility\Calculation
Python: MedianBinWidth.py (last modified: 2018-10-05)