re-exported rl_algs - fixed problems with serialization test and test_cartpole

baselines/common/tests/test_serialization.py

@@ -43,7 +43,7 @@ def test_serialization(learn_fn, network_fn):
             return 
 
     env = DummyVecEnv([lambda: MnistEnv(10, episode_len=100)])
-    ob = env.reset()
+    ob = env.reset().copy()
     learn = get_learn_function(learn_fn)
 
     kwargs = {}
@@ -51,24 +51,25 @@ def test_serialization(learn_fn, network_fn):
     kwargs.update(learn_kwargs[learn_fn])
 
 
-    learn = partial(learn, env=env, network=network_fn, seed=None, **kwargs)
+    learn = partial(learn, env=env, network=network_fn, seed=0, **kwargs)
 
     with tempfile.TemporaryDirectory() as td:
         model_path = os.path.join(td, 'serialization_test_model')
 
         with tf.Graph().as_default(), make_session().as_default():
-            model = learn(total_timesteps=100, seed=0)
+            model = learn(total_timesteps=100)
             model.save(model_path)
             mean1, std1 = _get_action_stats(model, ob)
             variables_dict1 = _serialize_variables()
 
         with tf.Graph().as_default(), make_session().as_default():
-            model = learn(total_timesteps=0, seed=0, load_path=model_path)
+            model = learn(total_timesteps=0, load_path=model_path)
             mean2, std2 = _get_action_stats(model, ob)
             variables_dict2 = _serialize_variables()
 
         for k, v in variables_dict1.items():
-            np.testing.assert_allclose(v, variables_dict2[k])
+            np.testing.assert_allclose(v, variables_dict2[k], atol=0.01,
+                err_msg='saved and loaded variable {} value mismatch'.format(k))
 
         np.testing.assert_allclose(mean1, mean2, atol=0.5)
         np.testing.assert_allclose(std1, std2, atol=0.5)

baselines/deepq/__init__.py

@@ -1,6 +1,6 @@
 from baselines.deepq import models  # noqa
 from baselines.deepq.build_graph import build_act, build_train  # noqa
-from baselines.deepq.deepq import learn, load  # noqa
+from baselines.deepq.deepq import learn, load_act  # noqa
 from baselines.deepq.replay_buffer import ReplayBuffer, PrioritizedReplayBuffer  # noqa
 
 def wrap_atari_dqn(env):

baselines/trpo_mpi/trpo_mpi.py

@@ -173,10 +173,6 @@ def learn(*,
     with tf.variable_scope("oldpi"):
         oldpi = policy(observ_placeholder=ob)
 
-    if load_path is not None:
-        pi.load(load_path)
-        oldpi.load(load_path)
-
     atarg = tf.placeholder(dtype=tf.float32, shape=[None]) # Target advantage function (if applicable)
     ret = tf.placeholder(dtype=tf.float32, shape=[None]) # Empirical return
 
@@ -247,6 +243,9 @@ def learn(*,
         return out
 
     U.initialize()
+    if load_path is not None:
+        pi.load(load_path)
+    
     th_init = get_flat()
     MPI.COMM_WORLD.Bcast(th_init, root=0)
     set_from_flat(th_init)