[Ifeffit] LCF and XES

[Matt Newville]

Hi Samy,

On Sun, Nov 4, 2018 at 2:47 AM Samy Ould-Chikh <samy.ouldchikh at kaust.edu.sa>
wrote:

> Dear,
>
>
>
> I’m trying to perform LCF using Athena for XES data (Kbeta 13) that has
> been normalized elsewhere.
>
> Combinatorics module is the main motivation.
>
>
>
> However, LCF module in Athena only use norm mu(E), so that the
> normalization for XAS data proceed and then it’s not possible to perform
> the fitting.
>
>
>
> Does anyone know a workaround to perform LCF for XES in Athena, or another
> alternative software as user friendly ?
>
>
Right now, I think Athena cannot do this easily.

The Larch XAS Viewer also has Linear Combo fitting with combinatorics.
This GUI (sort of like Athena) assumes that the data is held in Groups and
that the suffix is either "norm", "dmude", or "chi".  But that is
essentially a naming convention and could be worked around, at least in
principle (or the GUI code could be changed pretty easily).

For Larch, the GUI more or less sets up and runs the larch command
`lincombo_fitall()`  (see
http://cars.uchicago.edu/xraylarch/xafs/xanes.html#_math.lincombo_fitall).
 That function can be run from a script, without the GUI.  FWIW, the
function documentation says that it must use array names of "norm" or
"dmude", but that's not true -- it can be any array name, but all the
groups must use the same naming convention.

Anyway, you could have a script  read in the data to be fitted and the data
for the candidate standards.  I don't think it would need to be very
involved.

I think there would be others interested in such handling of XES data too.
I think it would not be too hard, but we might need some established
conventions for making sure that data was normalized and for naming things.

--Matt
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://millenia.cars.aps.anl.gov/pipermail/ifeffit/attachments/20181104/e774cbe8/attachment.html>