OscervaME
/
stable-baselines3-contrib-sacd
2
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Improve coverage

This commit is contained in:
Antonin RAFFIN 2020-10-12 20:17:33 +02:00
parent 7609c87e84
commit 5d7b79d41a
3 changed files with 6 additions and 16 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
sb3_contrib/tqc/policies.py
View File

@ -183,18 +183,6 @@ class Actor(BasePolicy):
def _predict(self, observation: th.Tensor, deterministic: bool = False) -> th.Tensor: def _predict(self, observation: th.Tensor, deterministic: bool = False) -> th.Tensor:
return self.forward(observation, deterministic) return self.forward(observation, deterministic)
def evaluate_actions(self, obs: th.Tensor, actions: th.Tensor) -> th.Tensor:
"""
Evaluate actions according to the current policy,
given the observations. Only useful when using SDE.
:param obs:
:param actions:
:return: log likelihood of taking those actions
"""
mean_actions, log_std, kwargs = self.get_action_dist_params(obs)
self.action_dist.proba_distribution(mean_actions, log_std, **kwargs)
return self.action_dist.log_prob(actions)
class Critic(BaseModel): class Critic(BaseModel):
""" """

4
tests/test_run.py
View File

@ -27,6 +27,8 @@ def test_n_critics(n_critics):
def test_sde(): def test_sde():
model = TQC( model = TQC(
"MlpPolicy", "Pendulum-v0", policy_kwargs=dict(net_arch=[64]), use_sde=True, learning_starts=100, verbose=1 "MlpPolicy", "Pendulum-v0", policy_kwargs=dict(net_arch=[64], sde_net_arch=[8]), use_sde=True, learning_starts=100, verbose=1
) )
model.learn(total_timesteps=500) model.learn(total_timesteps=500)
model.policy.reset_noise()
model.policy.actor.get_std()

6
tests/test_save_load.py
View File

@ -177,17 +177,17 @@ def test_set_env(model_class):
# create model # create model
model = model_class("MlpPolicy", env, policy_kwargs=dict(net_arch=[16])) model = model_class("MlpPolicy", env, policy_kwargs=dict(net_arch=[16]))
# learn # learn
model.learn(total_timesteps=500) model.learn(total_timesteps=300)
# change env # change env
model.set_env(env2) model.set_env(env2)
# learn again # learn again
model.learn(total_timesteps=500) model.learn(total_timesteps=300)
# change env test wrapping # change env test wrapping
model.set_env(env3) model.set_env(env3)
# learn again # learn again
model.learn(total_timesteps=500) model.learn(total_timesteps=300)
@pytest.mark.parametrize("model_class", MODEL_LIST) @pytest.mark.parametrize("model_class", MODEL_LIST)