Michal
Michal
I've got a prototype of something like that in [m93a/demonstration-general-fit-uncertainty](https://github.com/m93a/demonstration-general-fit-uncertainty) (try it [online](http://michal.grno.cz/mff/general-fit/)). I needed that for a school project. It uses TypeScript, Plotly.js and React but could be simplified...
@jobo322 Hey, My primary sources were these: [Numerical Recipes in FORTRAN 77, Chapter 15](https://websites.pmc.ucsc.edu/%7Efnimmo/eart290c_17/NumericalRecipesinF77.pdf#page=680&zoom=auto,-187,701) [Wikipedia: Levenberg-Marquardt](https://en.wikipedia.org/wiki/Levenberg–Marquardt_algorithm) [Damping–undamping strategies for the Levenberg–Marquardt nonlinear least-squares method](https://www.ssl.berkeley.edu/~mlampton/DampUndamp.pdf) [Wikipedia: Nonlinear regression](https://en.wikipedia.org/wiki/Nonlinear_regression) Plus some basic...
Now the only thing left is to add new tests and evaluate why the old ones fail.
Hey @jobo322, Sadly I don't know yet. The first thing to do would be to make a 3D plot of the function `sumOfSquaredResiduals(a, b)` and explore the parameter space of...
> I'd like to break it down into smaller bites I don't have an opinion on this, we'd better ask @targos 🙂 Should I push all the minor fixes we...
Sorry, I didn't have much time because of school and the other project that I'm working on (which will need this repo as a dependency but we're not quite there...
Hey, I kinda forgot about this PR too. I might get into it after the end of our exam period at some time in February. If I forget again, please...
@PlugaruT Bump! Any opinion on the PR? I've been using it for a month on my machine and it works like a charm!
I mean weighted as in "Weighted Total Least Squares", which apart from values for `x` and `y` needs their standard deviations. When computing the regression, it gives more weight to...
Okay, maybe it's not called Weighted Total Least Squares, my bad... 😀 I'm a layman in statistics – otherwise I'd have implemented an error-in-variables version of LM myself – but...