Re: [fityk-users] Fityk and Raman Spectra

[Marcin Wojdyr]

Could you write in more details about what you did before Fit > Run?
Maybe with a link to screenshot?

Marcin

On Wed, 23 Sep 2020 at 20:38, FitykNewuser <bs...@glcplus.com> wrote:
>
> I am attempting to use fityk to Lorentzian fit Raman spectra rather then using Origin since Origin is very expensive. The two issues I am running into is that fityk will not find the second order peaks on my Raman spectra, and when I select Run Fit, it gives me an error of "Trying to reverse singular matrix. Column 0 is zeroed." I'm not sure how to get around that. Please help! Thank you
>
> --
> http://groups.google.com/group/fityk-users
> ---
> You received this message because you are subscribed to the Google Groups "fityk-users" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to fityk-users...@googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/fityk-users/990a03f9-bb6b-4e1e-9d6e-602c60d85ac9n%40googlegroups.com.

[Matthew Graneri]

Hi there,

Sorry to resurrect this post, but I just came across the same problem. I am trying to fit several Voigt curves to my spectra, but I am getting an error saying 'Error: Trying to reverse singular matrix. Column 0 is zeroed.'

Unlike the original poster, I had fitted several guess peaks to the spectrum (not very good ones; the GUI didn't show me what was happening while I was dragging the cursor) and then pressed the 'start fitting' button, and got the error. I then tried Fit > Run, but that didn't work either.

I've attached a screenshot below to give you an idea of what I'm talking about. If there's any other information you need, just ask! I could really use any help that anyone can give me...

Regards,

Matthew

[Matthew Graneri]

I figured out the issue just after I sent this message. Sorry!

[Marek Tuhý]

Hi Matthew, could you please let me know how did you solve this? :) Struggling with the same issue. Thanks

Marek

Dne pondělí 9. ledna 2023 v 1:26:48 UTC+11 uživatel Matthew Graneri napsal:

[Matthew Graneri]

Hi Marek,

From memory, I think the issue was that I had some data points which were not actually data. I think they either appeared as 'nan' 'nan' (for x and y) in the CSV, or they were just extra empty lines. Because they weren't numbers, even though the graph would display, there was a point that didn't make sense to the program and so it couldn't fit the functions (you can tell if you use the line between data points option).

So my advice would be to check that there are no empty data points, or any other strange artefacts in your numbers. Go to Data > Table in the menu bar to find out.

Good luck!

Regards,

Matthew