def play_game(is_learning):
game.setup(1)
rAll = 0
d = False
j = 0
race_timer = 0
is_race_over = False
game.players[0].racers[1].draw_hand()
while j < 99:
j += 1
# choose an action by greedily (with noise) picking from Q table
s = game.current_observation()
a1 = RL.choose_action(str(s))
game.players[0].racers[1].select_move(a1)
game.players[0].racers[0].draw_hand()
s2 = game.current_observation()
a2 = RL.choose_action(str(s2))
game.players[0].racers[0].select_move(a2)
result = game.make_move()
if not is_learning:
racers = []
for player in game.players:
racers = [*racers, *player.racers]
racers.sort()
racers.reverse()
game.print_cards()
game.draw_course(racers)
if result:
print(j, 'turns taken')
r = 0
if result:
r = 1
s3 = 'terminal'
else:
game.players[0].racers[1].draw_hand()
s3 = game.current_observation()
if is_learning:
RL.learn(str(s2), a2, r, str((s3)))
RL.learn(str(s), a1, r, str((s2)))
rAll += r
if result:
break
rList.append(rAll)
import game
print('Random selection, 2 player game')
game.setup(2)
counter = 0
race_timer = 0
race_over = False
while not race_over:
if counter == 5:
counter = 0
game.print_cards()
for player in game.players:
for racer in player.racers:
racer.draw_hand()
racer.select_move(0)
race_over = game.make_move()
print('Next Round')
counter += 1
race_timer += 1
# sleep(2)
game.print_cards()
print('Finished in', race_timer, 'turns')
def test_clean_print_cards(self):
self.bob.add_card(Card('9', 'hearts'))
expected = '{}: {}'.format(self.bob.name, "Your hand: ['9 of Hearts']")
reality = game.print_cards(self.bob)
self.assertEqual(expected,reality)
def play_game(is_learning):
game.setup(2)
if not is_learning:
racers = []
for player in game.players:
racers = [*racers, *player.racers]
racers.sort()
racers.reverse()
game.draw_course(racers)
rAll = 0
d = False
j = 0
race_timer = 0
is_race_over = False
while j < 99:
j += 1
s, a1 = play_racer_RL(game.players[0].racers[1], is_learning)
s2, a2 = play_racer_RL(game.players[0].racers[0], is_learning)
play_racer_max(game.players[1].racers[0])
play_racer_max(game.players[1].racers[1])
if not is_learning:
game.print_cards_for_racer(game.players[1].racers[1])
game.print_cards_for_racer(game.players[1].racers[0])
result = game.make_move()
if not is_learning:
racers = []
for player in game.players:
racers = [*racers, *player.racers]
racers.sort()
racers.reverse()
game.draw_course(racers)
if result:
print(result.name, ' won,', j, 'turns taken')
r = 0
if result and (result == game.players[0].racers[0]
or result == game.players[0].racers[1]):
r = 1
elif result:
r = -1
s3 = game.current_observation()
if is_learning:
RL.store_transition(s, a1, r, s2)
RL.store_transition(s2, a2, r, s3)
if (counter > 500) and (counter % 10 == 0):
RL.learn()
rAll += r
if result:
break
if rAll == 0:
racers = []
for player in game.players:
racers = [*racers, *player.racers]
racers.sort()
racers.reverse()
game.print_cards()
game.draw_course(racers)
print(j)
print()
pass
rList.append(rAll)