This is a project where we attempt to make a cool chopsticks player
Outline:
Note all strategies are now mod n compatible, but we do not yet have good AIs
v0.0 - most basic possible player playing against player or computer. Records all player moves.
v1.0 - Records player moves. Windows only open picture, breaks on other os. v1.1 - i/o from voice and microphone. Incredibly Windows only breaks on other os. v1.2 - partially broken tkinter attempt v1.3 - cross platform speech, and pictures.
v2.0 - genetic strategy updating (no graphics or audio) v2.1 - genetic seeded with human data v2.2 - recursion delver v2.3 - genetic facing recursion delver v2.4 - my own speech v2.5 - my own speech recognition v2.6 - explore hyper parameters for genetic with grid search or something
v3.0 - suicide option v3.5 - preservation option
v4.0 - pip installable program with basic features and a good AI v4.1 - pip installable with ALL features
Conventions:
A state is an array with state = [ player_hand1, player_hand2, enemy_hand1, enemy_hand2 ] Note different players will observe this differently Note by convention state[0] <= state[1] and state[2] <= state[3] wlog for unique hands
Specific problems:
Add my speech recognition and cooler speaking with NNs better strategy