def evaluate(): wins = [0, 0] for i in range(100): player1 = randomPlayer() player2 = nTuplesSystematic() winner = game2.play(othello.game(), game2.player_epsilon(lambda x: player1.play_next_move(x)),game2.player_epsilon(lambda x: player2.play_next_move(x)), False) if winner == 1: wins[0] += 1 elif winner == 2: wins[1] += 1 winner = game2.play(othello.game(),game2.player_epsilon(lambda x: player2.play_next_move(x)), game2.player_epsilon(lambda x: player1.play_next_move(x)), False) if winner == 2: wins[0] += 1 elif winner == 1: wins[1] += 1 print wins
self.fin_val = [] def reset_without_train(self): self.moveb = 0 self.movew = 0 self.last_vb = 0 self.last_vw = 0 self.fin_v = [] self.fin_val = [] if __name__ == "__main__": """ Creates a main player """ playernew = nn() nTuplesSystematicObject = nts.nTuplesSystematic() game2.play(othello.game(), game2.player(lambda x: playernew.play_move(x)),game2.player(lambda x: nTuplesSystematicObject.play_next_move(x)), True) playernew.reset_without_train() time.sleep(5) k = 100 for i in range(k): print(i) game2.play(othello.game(), game2.player(lambda x: playernew.play_move(x,0.3)),game2.player(lambda x: playernew.play_move(x,0.3)), False) playernew.reset() wins = [0, 0] for i in range(100): winner = game2.play(othello.game(), game2.player_epsilon(lambda x: playernew.play_move(x)),game2.player_epsilon(lambda x: nTuplesSystematicObject.play_next_move(x)), False) if winner == 1: wins[0] += 1 elif winner == 2: