12
$\begingroup$

Now that Formulize / Eureqa started charging $2500 a year for using it and having crippled the trial version, does anyone know of any replacements that can do similar things like find an equation given data points?

It was free, but is now out of the price range of anyone that's not a company or school.

(it's a symbolic regression software)

Thanks

Improve this question
$\endgroup$
4
  • $\begingroup$ @Paul i use to input an array into formulize which would create periodic equations with 8000-44100 variables, then i would plug the equations back into octave and create several different periodic signals from the equations. $\endgroup$
    – Rick T
    Oct 10 '14 at 1:24
  • $\begingroup$ migrate to stats.stackexchange.com ? see algorithms-for-automatic-model-selection there. $\endgroup$
    – denis
    Feb 23 '15 at 17:08
  • $\begingroup$ I don't think that we need to migrate the question. It is on-topic here. $\endgroup$
    – nicoguaro ♦
    Feb 25 '15 at 5:41
  • $\begingroup$ I am impressed by how consistent different versions of the software are... both the 32 and 64 bit versions will crash within 2 seconds of trying to load a dataset of 4 triplets. Really impressed... Of course I only tried versions 1.82 AND 1.92, others might work ;) $\endgroup$
    – kiith10
    Jun 23 at 20:22
7
$\begingroup$

I wrote a Python package called PyPGE.

PyPGE is a Symbolic Regression implementation based on Prioritized Grammar Enumeration (1), not Evolutionary or Genetic Programming. It produces a deterministic Symbolic Regression algorithm.

(1) Worm, Tony, and Kenneth Chiu. "Prioritized grammar enumeration: symbolic regression by dynamic programming." Proceedings of the 15th annual conference on Genetic and evolutionary computation. ACM, 2013. GitHub: http://github.com/verdverm/pypge

Improve this answer
$\endgroup$
4
$\begingroup$

After a cursory google search on the subject, it appears that "symbolic regression" is a problem that lends itself greatly to stochastic optimization algorithms like genetic programming (GP). It is conceivable that you should look for an open source genetic programming library with modules specifically for symbolic regression, such as DEAP (Distributed Evolutionary Algorithms in Python).

Improve this answer
$\endgroup$
4
$\begingroup$

I once started writing anopen source version of Eureqa in Java. The project has limited capabilities but it implements the fitness function described in [1] and couple optimizations mentioned by the authors in other publications (e.g., searching for solutions in Pareto front).

Link: https://github.com/pkoperek/hubert

[1] Schmidt, Michael, and Hod Lipson. "Distilling free-form natural laws from experimental data." Science 324.5923 (2009): 81-85. DOI:10.1126/science.1165893

Improve this answer
$\endgroup$
3
$\begingroup$

I found the gramEvol R package flexible and easy to use. They have a small tutorial in which they rederive Kepler's third law from data.

Note that it relies on Genetic Programmic for its optimisation and thus might return different results if you run it twice.

Improve this answer
$\endgroup$
2
$\begingroup$

There is also a package for R called rgp. Visit this link.

https://cran.r-project.org/web/packages/rgp/index.html

I have not used rgp as I have only begun to use R seriously but it seemed like a good lead. I have another one for you that looks really promising but I have a mac and cannot use it:

http://dev.heuristiclab.com/wiki/AdditionalMaterial/ECML-PKDD

Improve this answer
$\endgroup$
1
  • 3
    $\begingroup$ Please elaborate on why you recommend the "rgp" package. $\endgroup$
    – Paul ♦
    Aug 11 '15 at 3:15
1
$\begingroup$

You can try TuringBot. It's similar to Eureqa but it uses Simulated Annealing for the optimization instead of a genetic algorithm.

Its free version runs on both Windows and Linux.

Improve this answer
$\endgroup$
1
  • $\begingroup$ Interesting but to get full functionality it cost $1000 dollars.... $\endgroup$
    – Rick T
    Feb 25 '20 at 6:15

Not the answer you're looking for? Browse other questions tagged or ask your own question.