class BotAgent:
def __init__(self, env, forget=False):
"""An agent based on a GOFAI bot."""
self.env = env
self.forget = forget
self.on_reset()
def on_reset(self):
self.bot = Bot(self.env, forget=self.forget)
def act(self, *args, **kwargs):
return {'action': self.bot.step()}
def analyze_feedback(self, reward, done):
pass
for run_no in range(options.num_runs):
level = level_dict[level_name]
mission_seed = options.seed + run_no
mission = level(seed=mission_seed)
expert = Bot(mission, forget=options.forget)
if options.verbose:
print(
'%s/%s: %s, seed=%d' %
(run_no + 1, options.num_runs, mission.surface, mission_seed))
try:
episode_steps = 0
while True:
action = expert.step()
obs, reward, done, info = mission.step(action)
total_reward += reward
episode_steps += 1
if done:
if reward > 0:
num_success += 1
total_steps += episode_steps
if reward <= 0:
print('FAILURE on %s, seed %d, reward %.2f' %
(level_name, mission_seed, reward))
break
except Exception as e:
print('FAILURE on %s, seed %d' % (level_name, mission_seed))