diff --git a/baselines/acer/acer_simple.py b/baselines/acer/acer_simple.py
index 3eaa6cc..0ca4fb9 100644
--- a/baselines/acer/acer_simple.py
+++ b/baselines/acer/acer_simple.py
@@ -243,7 +243,7 @@ class Runner(object):
             mb_mus.append(mus)
             mb_dones.append(self.dones)
             obs, rewards, dones, _ = self.env.step(actions)
-            # states information for statefull predictors like LSTM
+            # states information for statefull models like LSTM
             self.states = states
             self.dones = dones
             self.update_obs(obs, dones)
@@ -260,7 +260,7 @@ class Runner(object):
 
         mb_dones = np.asarray(mb_dones, dtype=np.bool).swapaxes(1, 0)
 
-        mb_masks = mb_dones # Used for statefull predictors like LSTM's to mask state when done
+        mb_masks = mb_dones # Used for statefull models like LSTM's to mask state when done
         mb_dones = mb_dones[:, 1:] # Used for calculating returns. The dones array is now aligned with rewards
 
         # shapes are now [nenv, nsteps, []]
diff --git a/baselines/acktr/kfac.py b/baselines/acktr/kfac.py
index 60af0ee..b420819 100644
--- a/baselines/acktr/kfac.py
+++ b/baselines/acktr/kfac.py
@@ -134,7 +134,7 @@ class KfacOptimizer():
         # check associated weights and bias for homogeneous coordinate representation
         # and check redundent factors
         # TO-DO: there may be a bug to detect associate bias and weights for
-        # forking layer, e.g. in inception predictors.
+        # forking layer, e.g. in inception models.
         for param in varlist:
             factorTensors[param]['assnWeights'] = None
             factorTensors[param]['assnBias'] = None