[proFit-list] Tabulating Multiple Fits in the 'points' mode

pro Fit support profit at quansoft.com
Sun Dec 5 06:46:13 CST 2010


Dear Simon

Unfortunately, the tabulate ('points') command is not designed to work with multi experiment fitting setup. We understand that such a command would be desirable, but we haven't decided yet on an easy to use design. The same applies to the parameter window, which doesn't allow to edit and display parameters for multiple experiments. The only place where this becomes visible is the multi-dim fitting window.

In order to use the fitted parameters from an multi experimental fit to calculate function values at the points measured and calculate the residuals, you basically need to go through all experiments, retrieve the fitted parameters of each experiment and set the function parameter (as visible in the parameter window) and then call the function for the data points measured.
You typically use such a statement for parameter retrieval and set up:
SetFunctionParam('DampedOscillation', 1, FitResult(fitResultObject fitRes, result fittedParameter, experiment 1, index1 1)); 

I append a full example including data, function and program to start the fit and get the residuals. The example has 3 experiments and 4 parameters with 2 of them being local to the experiments.

Kind regards, Peter

Peter Amrhein
Quantum Soft




On 30.11.2010, at 11:58, Lindhoud, Simon wrote:

> Dear pro Fit-users,
> 
> I am using pro Fit to carry out global fits of up to 9 data sets that I want to fit to variables that are unique to each data set, but also to a number of shared variables. The fitting script I use, therefore has multiple outputs (I use the multiple fit option). When I have done the fit of the data, I want to plot the data and the residuals of the data. To get the residuals of the fit I use the tabulate ('points') option. It turns out that I have to tabulate each each individual output, which makes it already quite tedious, however, things get worse since pro Fit apparently only tabulates the output of the fit against the last data set that I have selected to plot. This makes me go through the same procedure of plotting a data set, selecting the correct output of the fit and tabulating the fit nine times.
> Am I doing something wrong? Is there a way to do this more efficiently?
> 
> Furthermore I noticed that the parameters window and the Multidimensional Fitting window do not display the same information when it comes to the active output.
> Greetings,
> 
> Simon Lindhoud
> 
> 
> 
> 
> _______________________________________________
> proFit-list mailing list
> proFit-list at quantum-soft.com
> http://quantum-soft.com/mailman/listinfo/profit-list_quantum-soft.com

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://quantum-soft.com/pipermail/profit-list_quantum-soft.com/attachments/20101205/ef5b99a7/attachment-0002.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Multi dimensional fitting.zip
Type: application/zip
Size: 78573 bytes
Desc: not available
URL: <http://quantum-soft.com/pipermail/profit-list_quantum-soft.com/attachments/20101205/ef5b99a7/attachment-0001.zip>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://quantum-soft.com/pipermail/profit-list_quantum-soft.com/attachments/20101205/ef5b99a7/attachment-0003.html>


More information about the proFit-list mailing list