def setup_method(self):
self.env = GridWorldEnv(desc='4x4')
self.policy = ScriptedPolicy(
scripted_actions=[2, 2, 1, 0, 3, 1, 1, 1, 2, 2, 1, 1, 1, 2, 2, 1])
self.algo = Mock(env_spec=self.env.spec,
policy=self.policy,
max_episode_length=16)
def other_envs():
descs = [
['SFFF', 'FFFF', 'FFFF', 'FFFF'],
['FFSF', 'FFFH', 'FHFH', 'HFFG'],
['FHSF', 'FFFH', 'FHFH', 'HFFG'],
['FHSF', 'FGFH', 'FHFH', 'HFFH'],
['SHFF', 'HHFF', 'FFFF', 'FFFF'],
]
return [GridWorldEnv(desc=desc) for desc in descs]
def envs():
descs = [
['SFFF', 'FHFH', 'FFFH', 'HFFG'],
['SFFF', 'FFFH', 'FHFH', 'HFFG'],
['SFFF', 'FFFH', 'FHFH', 'FFFG'],
['SFFF', 'FFFF', 'FFFF', 'FFFF'],
['SHFF', 'HHFF', 'FFFF', 'FFFF'],
]
return [GarageEnv(GridWorldEnv(desc=desc)) for desc in descs]
def other_envs():
descs = [
['FFFS', 'FHFH', 'FFFH', 'HFFG'],
['FFSF', 'FFFH', 'FHFH', 'HFFG'],
['FFFF', 'FFSH', 'FHFH', 'FFFG'],
['FFFF', 'FFFF', 'FSFF', 'FFFF'],
['HHFF', 'HHHF', 'HSHF', 'HHHF'],
]
return [GarageEnv(GridWorldEnv(desc=desc)) for desc in descs]
class TestRaySamplerTF():
"""
Uses mock policy for 4x4 gridworldenv
'4x4': [
'SFFF',
'FHFH',
'FFFH',
'HFFG'
]
0: left
1: down
2: right
3: up
-1: no move
'S' : starting point
'F' or '.': free space
'W' or 'x': wall
'H' or 'o': hole (terminates episode)
'G' : goal
[2,2,1,0,3,1,1,1,2,2,1,1,1,2,2,1]
"""
def setup_method(self):
self.env = GridWorldEnv(desc='4x4')
self.policy = ScriptedPolicy(
scripted_actions=[2, 2, 1, 0, 3, 1, 1, 1, 2, 2, 1, 1, 1, 2, 2, 1])
self.algo = Mock(env_spec=self.env.spec,
policy=self.policy,
max_episode_length=16)
def teardown_method(self):
self.env.close()
def test_ray_batch_sampler(self, ray_local_session_fixture):
del ray_local_session_fixture
assert ray.is_initialized()
workers = WorkerFactory(
seed=100, max_episode_length=self.algo.max_episode_length)
sampler1 = RaySampler(workers, self.policy, self.env)
sampler1.start_worker()
sampler1.shutdown_worker()
def env():
return GridWorldEnv(desc='4x4')
def env():
return GarageEnv(GridWorldEnv(desc='4x4'))