def test_mutliproc_single_proc_equivalency():
env = gym.make("CartPole-v1")
num_envs = 3
# uses single threaded vector environment
venv1 = concat_vec_envs_v1(env, num_envs, num_cpus=0)
# uses multiprocessing vector environment
venv2 = concat_vec_envs_v1(env, num_envs, num_cpus=4)
check_vec_env_equivalency(venv1, venv2)
def test_multiagent_mutliproc_single_proc_equivalency():
env = simple_spread_v2.parallel_env(max_cycles=10)
env = pettingzoo_env_to_vec_env_v1(env)
num_envs = 3
# uses single threaded vector environment
venv1 = concat_vec_envs_v1(env, num_envs, num_cpus=0)
# uses multiprocessing vector environment
venv2 = concat_vec_envs_v1(env, num_envs, num_cpus=4)
check_vec_env_equivalency(venv1, venv2)
def create_envs(self,
n_envs: int,
eval_env: bool = False,
no_log: bool = False) -> VecEnv:
env = pistonball_v5.parallel_env()
env = ss.color_reduction_v0(env, mode="B")
env = ss.resize_v0(env, x_size=84, y_size=84, linear_interp=True)
env = ss.frame_stack_v1(env, 3)
env = ss.pettingzoo_env_to_vec_env_v1(env)
print(n_envs)
env = ss.concat_vec_envs_v1(env,
n_envs,
num_cpus=4,
base_class="stable_baselines3")
env = VecMonitor(env)
env = self._maybe_normalize(env, eval_env)
if is_image_space(
env.observation_space) and not is_image_space_channels_first(
env.observation_space):
if self.verbose > 0:
print("Wrapping into a VecTransposeImage")
env = VecTransposeImage(env)
return env
def test_multiproc_buffer():
num_envs = 2
env = gym.make("CartPole-v1")
env = concat_vec_envs_v1(env, num_envs, num_cpus=2)
obss = env.reset()
for i in range(55):
actions = [env.action_space.sample() for i in range(env.num_envs)]
# Check we're not passing a thing that gets mutated
keep_obs = copy.deepcopy(obss)
new_obss, rews, dones, infos = env.step(actions)
assert hash(str(keep_obs)) == hash(str(obss))
obss = new_obss
def test_good_vecenv():
num_envs = 2
env = simple_spread_v2.parallel_env()
max_num_agents = len(env.possible_agents) * num_envs
env = pettingzoo_env_to_vec_env_v1(env)
env = concat_vec_envs_v1(env, num_envs)
obss = env.reset()
for i in range(55):
actions = [env.action_space.sample() for i in range(env.num_envs)]
# Check we're not passing a thing that gets mutated
keep_obs = copy.deepcopy(obss)
new_obss, rews, dones, infos = env.step(actions)
assert hash(str(keep_obs)) == hash(str(obss))
assert len(new_obss) == max_num_agents
assert len(rews) == max_num_agents
assert len(dones) == max_num_agents
assert len(infos) == max_num_agents
# no agent death, only env death
if any(dones):
assert all(dones)
obss = new_obss
def image_transpose(env):
if is_image_space(env.observation_space) and not is_image_space_channels_first(
env.observation_space
):
env = VecTransposeImage(env)
return env
env = pistonball_v5.parallel_env()
env = ss.color_reduction_v0(env, mode="B")
env = ss.resize_v0(env, x_size=84, y_size=84)
env = ss.frame_stack_v1(env, 3)
env = ss.pettingzoo_env_to_vec_env_v1(env)
env = ss.concat_vec_envs_v1(env, n_envs, num_cpus=1, base_class="stable_baselines3")
env = VecMonitor(env)
env = image_transpose(env)
eval_env = pistonball_v5.parallel_env()
eval_env = ss.color_reduction_v0(eval_env, mode="B")
eval_env = ss.resize_v0(eval_env, x_size=84, y_size=84)
eval_env = ss.frame_stack_v1(eval_env, 3)
eval_env = ss.pettingzoo_env_to_vec_env_v1(eval_env)
eval_env = ss.concat_vec_envs_v1(
eval_env, 1, num_cpus=1, base_class="stable_baselines3"
)
eval_env = VecMonitor(eval_env)
eval_env = image_transpose(eval_env)
eval_freq = int(n_timesteps / n_evaluations)
import shutil
import traci
if __name__ == '__main__':
RESOLUTION = (3200, 1800)
env = sumo_rl.grid4x4(use_gui=True, out_csv_name='outputs/grid4x4/ppo_test', virtual_display=RESOLUTION)
max_time = env.unwrapped.env.sim_max_time
delta_time = env.unwrapped.env.delta_time
print("Environment created")
env = ss.pettingzoo_env_to_vec_env_v1(env)
env = ss.concat_vec_envs_v1(env, 2, num_cpus=1, base_class='stable_baselines3')
env = VecMonitor(env)
model = PPO("MlpPolicy",
env,
verbose=3,
gamma=0.95,
n_steps=256,
ent_coef=0.0905168,
learning_rate=0.00062211,
vf_coef=0.042202,
max_grad_norm=0.9,
gae_lambda=0.99,
n_epochs=5,
clip_range=0.3,
batch_size=256,
def test_inital_state_dissimilarity():
env = gym.make("CartPole-v1")
venv = concat_vec_envs_v1(env, 2)
observations = venv.reset()
assert not np.equal(observations[0], observations[1]).all()
def test_gym_supersuit_equivalency():
env = gym.make("MountainCarContinuous-v0")
num_envs = 3
venv1 = concat_vec_envs_v1(env, num_envs)
venv2 = gym_vec_env_v0(env, num_envs)
check_vec_env_equivalency(venv1, venv2)
env = pistonball_v6.parallel_env(
n_pistons=20,
time_penalty=-0.1,
continuous=True,
random_drop=True,
random_rotate=True,
ball_mass=0.75,
ball_friction=0.3,
ball_elasticity=1.5,
max_cycles=125,
)
env = ss.color_reduction_v0(env, mode="B")
env = ss.resize_v1(env, x_size=84, y_size=84)
env = ss.frame_stack_v1(env, 3)
env = ss.pettingzoo_env_to_vec_env_v1(env)
env = ss.concat_vec_envs_v1(env, 8, num_cpus=4, base_class="stable_baselines3")
model = PPO(
CnnPolicy,
env,
verbose=3,
gamma=0.95,
n_steps=256,
ent_coef=0.0905168,
learning_rate=0.00062211,
vf_coef=0.042202,
max_grad_norm=0.9,
gae_lambda=0.99,
n_epochs=5,
clip_range=0.3,
batch_size=256,
)
def test_multi_threaded_concatenate():
env = make_env()
env = pettingzoo_env_to_vec_env_v1(env)
env = concat_vec_envs_v1(env, 2, num_cpus=2)
dict_vec_env_test(env)
def test_env_is_wrapped_true():
env = gym.make("MountainCarContinuous-v0")
env = frame_skip_gym(env, 4)
num_envs = 3
venv1 = concat_vec_envs_v1(env, num_envs)
assert venv1.env_is_wrapped(frame_skip_gym) == [True] * 3
def test_env_is_wrapped_pettingzoo():
env = simple_spread_v2.parallel_env()
venv1 = pettingzoo_env_to_vec_env_v1(env)
num_envs = 3
venv1 = concat_vec_envs_v1(venv1, num_envs)
assert venv1.env_is_wrapped(frame_skip_gym) == [False] * 9