def step(self, action):
obs, reward, done, info = self.env.step(action)
if self.mode == 'Bob':
import ipdb
ipdb.set_trace()
#First visit done for Bob
if np.allclose(reward, 0.):
done = True
info['is_success'] = True
if info.get('TimeLimit.truncated'):
del info['TimeLimit.truncated']
return obs, reward, done, info
elif self.mode == 'Alice':
import ipdb
ipdb.set_trace()
info = AttrDict(info)
self.total_rewards += reward
if done:
done = False
info.done_observation = obs
#info.terminal_state = True
if info.get('TimeLimit.truncated'):
done = True
info.terminal_state = False
info.episodic_return = self.total_rewards
self.total_rewards = 0
else:
info.terminal_state = False
info.episodic_return = None
return obs, reward, done, info
def step(self, action):
obs, reward, done, info = self.env.step(action)
info = AttrDict(info)
self.total_rewards += reward
if done:
info.done_observation = obs
info.terminal_state = True
if info.get('TimeLimit.truncated'):
info.terminal_state = False
info.episodic_return = self.total_rewards
self.total_rewards = 0
else:
info.terminal_state = False
info.episodic_return = None
return obs, reward, done, info
def __init__(
self,
module_list: Iterable, # list of mrl.Modules (possibly nested)
config: AttrDict): # hyperparameters and module settings
self.config = config
parent_folder = config.parent_folder
assert parent_folder, "Setting the agent's parent folder is required!"
self.agent_name = config.get(
'agent_name') or 'agent_' + short_timestamp()
self.agent_folder = os.path.join(parent_folder, self.agent_name)
load_agent = False
if os.path.exists(self.agent_folder):
print('Detected existing agent! Loading agent from checkpoint...')
load_agent = True
else:
os.makedirs(self.agent_folder, exist_ok=True)
self._process_experience_registry = [
] # set of modules which define _process_experience
self._optimize_registry = [] # set of modules which define _optimize
self.config.env_steps = 0
self.config.opt_steps = 0
module_list = flatten_modules(module_list)
self.module_dict = AttrDict()
for module in module_list:
assert module.module_name
setattr(self, module.module_name, module)
self.module_dict[module.module_name] = module
for module in module_list:
self._register_module(module)
self.training = True
if load_agent:
self.load()
print('Successfully loaded saved agent!')
else:
self.save()