.. algorithm:: .. summary:: .. relatedalgorithms:: .. properties:: Description ----------- The inputs to this algorithm are single-spectra workspaces containing polarization efficiencies. They are combined and interpolated if necessary to form a valid matrix workspace. The spectra of the output workspace are labeled with the names of the corresponding input properties. Usage ----- .. testcode:: Example # Create input workspaces which can have different sizes ws1 = CreateWorkspace([1, 2, 3], [1, 1]) ws2 = CreateWorkspace([2, 3, 4, 5], [1, 1, 1]) # Combine them in a single workspace efficiencies = JoinISISPolarizationEfficiencies(Pp=ws1, Ap=ws2) print('Number of spectra = {}'.format(efficiencies.getNumberHistograms())) print('Number of bins = {}'.format(efficiencies.blocksize())) print('Label of first spectrum: {}'.format(efficiencies.getAxis(1).label(0))) print('Label of second spectrum: {}'.format(efficiencies.getAxis(1).label(1))) Output: .. testoutput:: Example Number of spectra = 2 Number of bins = 3 Label of first spectrum: Pp Label of second spectrum: Ap .. categories:: .. sourcelink::