[Ifeffit] Project files and fits

[Stefano Ciurli]

Hello Bruce and Norbert,
thanks a lot for your response! I feel less lonely.... :-)

>That is a splendid idea.  In fact it is such a good idea that I have
>been working hard for the past three weeks on exactly that.

:-))

>Actually, my plan is a bit more ambitious.  Artemis will save *every*
>fit you perform along with its log file and some other information
>into the project.

This is another question I have planned to ask and never got there:
let's say that I want to try different inp files (different models) 
to fit a single spectrum in artemis. What I currently do is that for 
every fit I save a project, and after closing it, I reload the 
spectrum and another input file and redo the feff calculation and the 
fit and so on. In this way I am producing a project file for each fit 
tried.
Of course I tried to work always in the same project, by loading the 
spectrum once and then loading several input files and for each 
running a feff calculation and a fit. However, I find that for each 
input file the program creates a set of amp, e0, delr and ss and 
calls these parameters with different names for each feff calculation 
(for example amp for feff0, amp_1 for feff1 etc). This coule be very 
nice indeed were it not for the fact that the program also keeps 
active the previous parameters, and I have all the times to skip 
them. If they are only four parameters, the time consumed is not an 
issue, but when I start having a larger number of paramers then it is 
quite annoying. Wouldn't it be better to set things so that everytime 
you load a new input file and a new feff calculation within the same 
project the previous parameters, used for the previously tried fit, 
were automatically skipped? Just a thought. If I have not been clear 
let me know I'll try to rephrase things.

>Each fit will get its own entry in the list of data
>and paths -- that means that you will be able to plot several fits
>along with the data in order to see what effect each change has on the
>quality of fit.

This would be even greater than it is already (:-)) if the point 
above is taken into consideration. Unless again I do things in a non 
correct way, which is possible if not likely.

>  And since each log file is saved, you can do things
>like generate reports on the log files to see how the R-factor,
>chi-square, and guess parameters have evolved with every change in the
>fitting model.

This is already doable, but I guess things can be improved, as we 
previously discussed.

>This is already available in a limited, buggy form in the new
>development code in the latest source tarball.  I am hoping to
>transition these features into the stable version by the end of the
>year.  Something to look foward to...

I do look forward to this! :-))

Ciao,
Stefano


-- 
____________________________________________

Stefano Ciurli
Professor of Chemistry
Department of Agro-Environmental Science and Technology
University of Bologna
Viale Giuseppe Fanin, 40
I-40127 Bologna
Italy
Phone:	+39-051-209-6204
Fax:	+39-051-209-6203

"Fatti non foste a viver come bruti,
ma per seguir virtute e canoscenza"
Dante Alighieri - Inferno - Canto XXVI

"Ihr seid bestimmt, nicht Tieren gleich zu leben,
Nein, Tugend zu erringen und Erkenntnis"
"Ye were not form'd to live the life of brutes,
But virtue to pursue and knowledge high"