Skip to content

microprediction/optimizer-elo-ratings

Repository files navigation

Behold a colab notebook that recommends a black-box optimizer for your objective function. It is powered by this repository, that assigns Elo ratings (json and html) to global derivative-free Python global optimization "strategies", using the HumpDay package.

Methodology

See the article for a detailed description.

What's a strategy?

A choice of python package (such as scipy.optimize, ax-platform, hyperopt, nevergrad, optuna, platypus, pymoo, pySOT or skopt) together with a fixing of additional choices (such as selection of method, choice of parameters etc).

Create your own shortlist

HumpDay contains scripts for choosing optimizers based on your own problems.

Contribute

Got a good objective function, or optimization strategy? Shove it in the HumpDay package and file a pull request, or at least suggest it in the issues.

About

Assigns Elo Ratings to Python Global Optimizers

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 4

  •  
  •  
  •  
  •