def make_env_and_successful_return(self, test):
env = ABC(
discrete=False,
episodic=self.episodic or test,
deterministic=test,
)
return env, 1
def make_env_and_successful_return(self, test):
env = ABC(
discrete=self.discrete,
episodic=self.episodic or test,
deterministic=test,
partially_observable=self.recurrent,
)
return env, 1
def make_env(process_idx, test):
size = 2
return ABC(
size=size,
discrete=discrete,
episodic=episodic or test,
partially_observable=self.recurrent,
deterministic=test,
)
def make_env(process_idx, test):
size = 2
return ABC(
size=size,
discrete=discrete,
episodic=True,
partially_observable=self.use_lstm,
deterministic=test,
)
def setUp(self):
self.env = ABC(deterministic=True)
self.obs_shape = self.env.observation_space.shape
self.model = torch.nn.Sequential(
torch.nn.Linear(self.obs_shape[0], 16),
torch.nn.ReLU(),
torch.nn.Linear(16, 16),
torch.nn.ReLU(),
torch.nn.Linear(16, 3),
)
def make_env():
return ABC(discrete=self.discrete, deterministic=test)
def make_env_and_successful_return(self, test):
return ABC(discrete=True, deterministic=test), 1
def make_env_and_successful_return(self, test):
return ABC(discrete=True,
partially_observable=True,
deterministic=test), 1
