def run(self):
sess = U.single_threaded_session()
sess.__enter__()
env = self.make_env()
workerseed = self.seed + 10000 * MPI.COMM_WORLD.Get_rank()
set_all_seeds(workerseed)
env.seed(workerseed)
pi = self.make_pi('pi%s' % os.getpid(), env.observation_space, env.action_space)
print('Worker %s - Running with seed %s' % (os.getpid(), workerseed))
while True:
self.event.wait()
self.event.clear()
command, weights = self.input.get()
if command == 'collect':
#print('Worker %s - Collecting...' % os.getpid())
pi.set_parameter(weights)
samples = self.traj_segment_generator(pi, env)
self.output.put((os.getpid(), samples))
elif command == 'exit':
print('Worker %s - Exiting...' % os.getpid())
env.close()
sess.close()
break
def run(self):
sess = U.single_threaded_session()
sess.__enter__()
env = self.make_env(self.index)
self.env = env
env.reset()
workerseed = self.seed + 10000 * MPI.COMM_WORLD.Get_rank()
set_all_seeds(workerseed)
env.seed(workerseed)
scope = 'pi%s' % os.getpid()
pi = self.make_pi(scope, env)
var = get_pi_trainable_variables(scope)
set_from_flat = U.SetFromFlat(var)
print('Worker %s - Running with seed %s' % (os.getpid(), workerseed))
while True:
self.event.wait()
self.event.clear()
command, weights = self.input.get()
if command == 'collect':
set_from_flat(weights)
#print('Worker %s - Collecting...' % os.getpid())
samples = self.traj_segment_generator(pi, env)
self.output.put((os.getpid(), samples))
elif command == 'exit':
print('Worker %s - Exiting...' % os.getpid())
env.close()
sess.close()
break
def _thunk():
_env = gym.make(env)
_env.seed(seed)
set_all_seeds(seed)
return _env
def _thunk():
_env = make_atari(env)
_env.seed(seed)
set_all_seeds(seed)
return wrap_deepmind(_env)
def _thunk():
env_rllab = Rllab2GymWrapper(env_rllab_class())
env_rllab.seed(seed)
set_all_seeds(seed)
return env_rllab
def seed(self, seed=0):
set_all_seeds(seed)