If I was doing this project I would do a monte carlo search instead of the much more common alpha-beta min-max search. Among other advantages, you don't actually have to write an evaluation function (although you can use one). Look it up on wikipedia.
Chess python software
Sort:
If the project actually requires writing an evaluation function, you can still make a minimal one, and use it to help choose the nodes to expand.
madratter7 wrote:
If I was doing this project I would do a monte carlo search instead of the much more common alpha-beta min-max search. Among other advantages, you don't actually have to write an evaluation function (although you can use one). Look it up on wikipedia.
Thank you for your response - I will research Monte carlo search. The project doesn't actually have any specific requirements - I can chose however I want to execute it, so there is a lot of flexibility allowed.
I have to make a chess software (player vs AI) as part of a school project - so clearly it doesn't have to be any match to stockfish etc. Just some basic AI which has some sort of rating function which calculates the strength of the position in the different variations and uses that to decide upon a move. Do any of you have any suggestions on how to get started? Perhaps any example code that you may have coded?
Any help is greatly appreciated