PEP-8 Fixes in algorithmic environment (#1382)

Remove trailing whitespaces.
Make line breaks adhere to 80 character limit (not all, but quite a few).
Remove unused imports.
Other miscellaneous PEP-8 fixes.

gym/envs/algorithmic/algorithmic_env.py

@@ -59,8 +59,8 @@ class AlgorithmicEnv(Env):
         self.last = 10
         # Cumulative reward earned this episode
         self.episode_total_reward = None
-        # Running tally of reward shortfalls. e.g. if there were 10 points to earn and
-        # we got 8, we'd append -2
+        # Running tally of reward shortfalls. e.g. if there were 10 points to
+        # earn and we got 8, we'd append -2
         AlgorithmicEnv.reward_shortfalls = []
         if chars:
             self.charmap = [chr(ord('A')+i) for i in range(base)]
@@ -78,7 +78,8 @@ class AlgorithmicEnv(Env):
         self.action_space = Tuple(
             [Discrete(len(self.MOVEMENTS)), Discrete(2), Discrete(self.base)]
         )
-        # Can see just what is on the input tape (one of n characters, or nothing)
+        # Can see just what is on the input tape (one of n characters, or
+        # nothing)
         self.observation_space = Discrete(self.base + 1)
         self.seed()
         self.reset()
@@ -170,10 +171,11 @@ class AlgorithmicEnv(Env):
             try:
                 correct = pred == self.target[self.write_head_position]
             except IndexError:
-                logger.warn("It looks like you're calling step() even though this "+
-                    "environment has already returned done=True. You should always call "+
-                    "reset() once you receive done=True. Any further steps are undefined "+
-                    "behaviour.")
+                logger.warn(
+                    "It looks like you're calling step() even though this "
+                    "environment has already returned done=True. You should "
+                    "always call reset() once you receive done=True. Any "
+                    "further steps are undefined behaviour.")
                 correct = False
             if correct:
                 reward = 1.0
@@ -209,12 +211,11 @@ class AlgorithmicEnv(Env):
         AlgorithmicEnv.reward_shortfalls.append(self.episode_total_reward - len(self.target))
         AlgorithmicEnv.reward_shortfalls = AlgorithmicEnv.reward_shortfalls[-self.last:]
         if len(AlgorithmicEnv.reward_shortfalls) == self.last and \
-          min(AlgorithmicEnv.reward_shortfalls) >= self.MIN_REWARD_SHORTFALL_FOR_PROMOTION and \
-          AlgorithmicEnv.min_length < 30:
+                min(AlgorithmicEnv.reward_shortfalls) >= self.MIN_REWARD_SHORTFALL_FOR_PROMOTION and \
+                AlgorithmicEnv.min_length < 30:
             AlgorithmicEnv.min_length += 1
             AlgorithmicEnv.reward_shortfalls = []
 
-
     def reset(self):
         self._check_levelup()
         self.last_action = None
@@ -264,7 +265,7 @@ class TapeAlgorithmicEnv(AlgorithmicEnv):
 
     def render_observation(self):
         x = self.read_head_position
-        x_str =      "Observation Tape    : "
+        x_str = "Observation Tape    : "
         for i in range(-2, self.input_width + 2):
             if i == x:
                 x_str += colorize(self._get_str_obs(np.array([i])), 'green', highlight=True)
@@ -278,6 +279,7 @@ class GridAlgorithmicEnv(AlgorithmicEnv):
     """An algorithmic env with a 2-d input grid."""
     MOVEMENTS = ['left', 'right', 'up', 'down']
     READ_HEAD_START = (0, 0)
+
     def __init__(self, rows, *args, **kwargs):
         self.rows = rows
         AlgorithmicEnv.__init__(self, *args, **kwargs)
@@ -316,7 +318,7 @@ class GridAlgorithmicEnv(AlgorithmicEnv):
 
     def render_observation(self):
         x = self.read_head_position
-        label =      "Observation Grid    : "
+        label = "Observation Grid    : "
         x_str = ""
         for j in range(-1, self.rows+1):
             if j != -1:

gym/envs/algorithmic/copy_.py

@@ -4,10 +4,10 @@ the output tape. http://arxiv.org/abs/1511.07275
 """
 from gym.envs.algorithmic import algorithmic_env
 
+
 class CopyEnv(algorithmic_env.TapeAlgorithmicEnv):
     def __init__(self, base=5, chars=True):
         super(CopyEnv, self).__init__(base=base, chars=chars)
 
     def target_from_input_data(self, input_data):
         return input_data
-

gym/envs/algorithmic/duplicated_input.py

@@ -5,6 +5,7 @@ http://arxiv.org/abs/1511.07275
 from __future__ import division
 from gym.envs.algorithmic import algorithmic_env
 
+
 class DuplicatedInputEnv(algorithmic_env.TapeAlgorithmicEnv):
     def __init__(self, duplication=2, base=5):
         self.duplication = duplication

gym/envs/algorithmic/repeat_copy.py

@@ -4,12 +4,13 @@ the output tape. http://arxiv.org/abs/1511.07275
 """
 from gym.envs.algorithmic import algorithmic_env
 
+
 class RepeatCopyEnv(algorithmic_env.TapeAlgorithmicEnv):
     MIN_REWARD_SHORTFALL_FOR_PROMOTION = -.1
+
     def __init__(self, base=5):
         super(RepeatCopyEnv, self).__init__(base=base, chars=True)
         self.last = 50
 
     def target_from_input_data(self, input_data):
         return input_data + list(reversed(input_data)) + input_data
-

gym/envs/algorithmic/reverse.py

@@ -2,11 +2,12 @@
 Task is to reverse content over the input tape.
 http://arxiv.org/abs/1511.07275
 """
-
 from gym.envs.algorithmic import algorithmic_env
 
+
 class ReverseEnv(algorithmic_env.TapeAlgorithmicEnv):
     MIN_REWARD_SHORTFALL_FOR_PROMOTION = -.1
+
     def __init__(self, base=2):
         super(ReverseEnv, self).__init__(base=base, chars=True, starting_min_length=1)
         self.last = 50

gym/envs/algorithmic/reversed_addition.py

@@ -1,7 +1,7 @@
 from __future__ import division
-import numpy as np
 from gym.envs.algorithmic import algorithmic_env
 
+
 class ReversedAdditionEnv(algorithmic_env.GridAlgorithmicEnv):
     def __init__(self, rows=2, base=3):
         super(ReversedAdditionEnv, self).__init__(rows=rows, base=base, chars=False)