n-p / RWO

Random Walking Optimization

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

What's new

  • Automatic detection of tuning variables.
  • It shows which values changed during the last iteration.
  • It's testing just small random samples before testing them thoroughly.

How it works?

This tuner is using the simple random walking for position-based tuning of stockfish chess engine (using centipawn values generated by it). It can be applied to other engines as well with some modifications.

Important notes:

Analyzed positions here has been processed by older developement version of stockfish. So be aware of it or you should create them again using analysis.py and latest development version of stockfish.

About

Random Walking Optimization

License:GNU General Public License v3.0


Languages

Language:Python 100.0%