def test_run(self):
env = Hanabi(agents=list('1234'), hand_size=4)
done = False
while not done:
env.render()
action = Actions.sample(hand_size=4, players=4, hints=env.info['hints'])
obs, reward, done, info = env.step(action)
def test_speed_runs(self):
env = Hanabi(agents=list('1234'), hand_size=4)
steps = 0
for episode in range(1, 1_001):
done = False
env.reset()
reward = 0
while not done:
action = Actions.sample(hand_size=4, players=4, hints=env.info['hints'])
obs, reward, done, info = env.step(action)
steps += 1
print(f"\rEpisode: {episode:4d}/1000, steps: {steps:6d}, reward: {reward}", end='')
from environment import Hanabi
from environment.utils.constants import Actions
if __name__ == '__main__':
env = Hanabi(agents=list('12'), )
# Run a single game loop
obs = env.reset()
done = False
while not done:
action = Actions.sample(hand_size=env.hand_size, players=len(env.players), hints=env.hints)
obs, reward, done, info = env.step(action)
env.render()
# Unpack an observations
for key, value in obs.items():
if key == 'turns log':
print(f"Key: {key} (unpacked dict)", end='\n\n')
for turn, value in value.items():
print(f"\t - Turn {turn}: {' '.join(map(str, value))}")
continue
print(f"Key: {key}\n{value}", end='\n\n')
def test_action_sample(self):
env = Hanabi(agents=list('1234'), hand_size=4)
env.render()
env.step(Actions.sample(hand_size=4, players=4, hints=env.info['hints']))
env.render()
def test_action_discard(self):
env = Hanabi(agents=list('1234'), hand_size=4)
env.render()
env.step(Actions.DISCARD(1))
env.render()
def test_action_inform_rank(self):
env = Hanabi(agents=list('1234'), hand_size=4)
env.render()
env.step(Actions.INFORM_RANK(2, 3))
env.render()
def test_action_inform_color(self):
env = Hanabi(agents=list('1234'), hand_size=4)
env.render()
env.step(Actions.INFORM_COLOR(2, 3))
env.render()
def test_action_play(self):
env = Hanabi(agents=list('1234'), hand_size=4)
env.render()
env.step(Actions.PLAY(1, 2))
env.render()