muD as a required input

[till-schertenleib]

Problem

Proposed solution

there are two issues:

1. why am I required to specify muD if I have a zscan file to determine muD experimentally?
2. I specified a muD of 1.0, why does it tell me that this is outside o fhte acceptable range for muD (0.5 to 6.0)?

I guessed the muD of 1.0, since I don't know the muD of my sample. muD should only be a requirement if I don't provide a zscan file... no?

[till-schertenleib]

Here is the input line and the error I got back.

$ labpdfproc 1.0 ts1000_635um_3_accum_0.xy -w 0.71073 -x tth -z ts1000_635um_zscan_exported.xy
Traceback (most recent call last):
File "C:\Users\tills\miniconda3\envs\labpdfproc\Scripts\labpdfproc-script.py", line 9, in
sys.exit(main())
^^^^^^
File "C:\Users\tills\miniconda3\envs\labpdfproc\Lib\site-packages\diffpy\labpdfproc\labpdfprocapp.py", line 169, in main
absorption_correction = compute_cve(input_pattern, args.mud, args.method)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\tills\miniconda3\envs\labpdfproc\Lib\site-packages\diffpy\labpdfproc\functions.py", line 267, in compute_cve
abdo_on_global_tth = cve_function(diffraction_data, mud)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\tills\miniconda3\envs\labpdfproc\Lib\site-packages\diffpy\labpdfproc\functions.py", line 213, in _cve_polynomial_interpolation
raise ValueError(
ValueError: mu*D is out of the acceptable range (0.5 to 6) for polynomial interpolation. Please rerun with a value within this range or specifying another method from ('brute_force', 'polynomial_interpolation').

[yucongalicechen]

Yeah you're right. I'll implement this issue #143 to fix this. Right now if you have a zscan file, muD will be replaced by whatever it fits for the file. So it looks like the zscan muD here is <0.5 or > 6.

[sbillinge]

Thanks for reporting Till. For <0.5 I don't think we want to throw a value error, we can just extrapolate or set to 0.

I guess this is teh second time someone wanted the fast interpolator to work for mud >6. We could offer this in principle. It just means calculating a few more curves, right? @till-schertenleib, how high do you want the MuD to go and still have the fast calculator work?

Another thought, on the error message, it is probably a good idea to print what the value of hte muD is so the user has some idea why it failed.

[till-schertenleib]

I think more often people will have muD < 0.5. My sample was a Ce MOF in a 0.6mm capillary. I was surprised that the muD was that low. for muD < 0, the correction might not be necessary, but probably wouldn't make the PDF worse. So we can make the interpolation work for such low muD. For the high boundary: we know that a muD of around 8 is too much to still get useful data (from our HfO2 data). 5.8 still gave decent results. So maybe we can make it work until 7?

…

On Thu, 6 Feb 2025 at 01:32, Simon Billinge ***@***.***> wrote: Thanks for reporting Till. For <0.5 I don't think we want to throw a value error, we can just extrapolate or set to 0. I guess this is teh second time someone wanted the fast interpolator to work for mud >6. We could offer this in principle. It just means calculating a few more curves, right? @till-schertenleib <https://github.com/till-schertenleib>, how high do you want the MuD to go and still have the fast calculator work? Another thought, on the error message, it is probably a good idea to print what the value of hte muD is so the user has some idea why it failed. — Reply to this email directly, view it on GitHub <#165 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/A2XYJRX3R32Q5B5CNRULVEL2OKUS3AVCNFSM6AAAAABWRIHZYKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDMMZYGM2DAOBRG4> . You are receiving this because you were mentioned.Message ID: ***@***.***>

-- Till Schertenleib EPFL, Switzerland Phone: +41 78 729 21 31

[yucongalicechen]

Closed by #168 and #170.
@till-schertenleib Could you try the new muD workflow when you get a chance? Thanks!
The CLI works as follows: labpdfproc zro2_mo.xy --mud 2.5 or labpdfproc zro2_mo.xy -z zscanfilepath. You can also type labpdfproc --gui for the gui interface. Let me know how it works! I'll also update the instructions in the docs and readme soon.