[Ifeffit] Output of data fit in k-space from current version of Artemis

[Bruce Ravel]

Hi Rosalie,

After spending some time with this, I agree that there is something
very wrong that manifests itself when you attempt to save the data+fit
file in k-space.

The executive summary is that there is a problem deep under the hood
with the k-grid that Artemis uses for the fitted function.  The fitted
function is not wrong -- but there is a confusion regarding the
k-grid.  For reasons of how Artemis does it's various chores, this is
manifested when saving the data+fit file in k, but it is not
manifested when making certain kinds of plots.  That is, as you
noticed, the plot looks sensible but the data+fit save file is incorrect.

I have a work-around for you while to try I continue to puzzle over
this one.  Download a copy of this small text file:

 
https://raw.githubusercontent.com/bruceravel/demeter/master/lib/Demeter/templates/fit/ifeffit/residual.tmpl

On your computer, overwrite this file:

   C:\strawberry\perl\site\lib\Demeter\templates\fit\ifeffit\residual.tmpl

preserving the file name.

Rerun your fit, then try saving the data+fit output file.

Let me know if that works for you.  Hopefully I will figure out the
underlying problem.  If not, it will be good to know if this crude fix
works for you.

Cheers,
B


On 05/26/2015 09:18 AM, Bruce Ravel wrote:
> On 05/25/2015 08:37 PM, Rosalie Hocking wrote:
>> If I go to the Artemis data window and go Data “save data and fit as” it
>> doesn’t seem to match up with the simulations on the window.
>
> Hi Rosalie,
>
> I don't see a problem.  In the submenu you mention, there are options
> for saving data and fit in k-space with various different k-weightings.
>   When I select the same k-weighting that I am currently using in the
> plot, save the file, then replot, the data -- it looks the same to me.
>
> I hate to be the jerk who says "have you considered this thing which is
> plainly obvious?" ... but ... are you sure that you are saving the data
> with the same k-weighting as the plot?
>
> Also, I just want to point out that the k-weighted data are in column 2
> of that file and the k-weighted fit is in column 3.
>
> B
>
>


-- 
  Bruce Ravel  ------------------------------------ bravel at bnl.gov

  National Institute of Standards and Technology
  Synchrotron Science Group at NSLS-II
  Building 535A
  Upton NY, 11973

  Homepage:    http://bruceravel.github.io/home/
  Software:    https://github.com/bruceravel
  Demeter:     http://bruceravel.github.io/demeter/