keithgw / wingspan

A Wingspan AI that uses reinforcement learning to learn to play the wingspan board game.
2 stars 1 forks source link

move all members of the game state into GameState #52

Closed keithgw closed 8 months ago

keithgw commented 8 months ago

To help BotPlayers keep track of the changing state, it is required that their known_missing_cards be updated every time the tray is refilled even when it is not their turn. That is, knowable information about the game state can change in between their turns, that cannot be then deduced when it is again their turn. To make this easier to manage, move all of the members of the game state into the GameState class, this includes:

keithgw commented 8 months ago

A dependency for #51

keithgw commented 8 months ago

The game should be able to play in the same way it does now, that is start from the beginning of the game and have each player take a turn until the game is over.

Modifying the play loop to allow play from within a phase of a turn will be addressed in #53

keithgw commented 8 months ago

54