search

justNo4b / Drofa

UCI chess engine
GNU General Public License v3.0
46 stars 7 forks source link

Implement new net (KP -> 8 -> 2). Train on bigger dataset also #215

Closed justNo4b closed 1 year ago

justNo4b commented 1 year ago

It is possible that P net would be better and good result is consequence of bigger data. Will check a bit later