diff --git a/setup.py b/setup.py
index db8f51bc98..3bae8f3040 100644
--- a/setup.py
+++ b/setup.py
@@ -33,4 +33,4 @@
       author='OpenAI',
       url='https://github.com/openai/stable_baselines',
       author_email='gym@openai.com',
-      version='0.1.7')
+      version='0.2.0')
diff --git a/stable_baselines/acer/acer_simple.py b/stable_baselines/acer/acer_simple.py
index 4b3193de98..9c8c67a835 100644
--- a/stable_baselines/acer/acer_simple.py
+++ b/stable_baselines/acer/acer_simple.py
@@ -155,8 +155,6 @@ def setup_model(self):
             else:
                 raise ValueError("Error: ACER does not work with {} actions space.".format(self.action_space))
 
-            self.n_batch = self.n_envs * self.n_steps
-
             self.graph = tf.Graph()
             with self.graph.as_default():
                 self.sess = tf_util.make_session(num_cpu=self.num_procs, graph=self.graph)
diff --git a/tests/test_identity.py b/tests/test_identity.py
index 18ee31560a..9698f18006 100644
--- a/tests/test_identity.py
+++ b/tests/test_identity.py
@@ -19,7 +19,7 @@
                    n_steps=1, replay_ratio=1).learn(total_timesteps=10000, seed=0),
     lambda e: ACKTR(policy=MlpPolicy, env=e, learning_rate=5e-4, n_steps=1).learn(total_timesteps=20000, seed=0),
     lambda e: DeepQ(policy=deepq_models.mlp([16]), batch_size=16, gamma=0.1,
-                    exploration_fraction=0.001, env=e).learn(total_timesteps=20000, seed=0),
+                    exploration_fraction=0.001, env=e).learn(total_timesteps=30000, seed=0),
     lambda e: PPO1(policy=MlpPolicy, env=e, lam=0.7,
                    optim_batchsize=16, optim_stepsize=1e-3).learn(total_timesteps=10000, seed=0),
     lambda e: PPO2(policy=MlpPolicy, env=e, learning_rate=1.5e-3,