Cleanup, removal of unmaintained code (#836)

* add dtype to Box

* remove board_game, debugging, safety, parameter_tuning environments

* massive set of breaking changes
- remove python logging module
- _step, _reset, _seed, _close => non underscored method
- remove benchmark and scoring folder

* Improve render("human"), now resizable, closable window.

* get rid of default step and reset in wrappers, so it doesn’t silently fail for people with underscore methods

* CubeCrash unit test environment

* followup fixes

* MemorizeDigits unit test envrionment

* refactored spaces a bit
fixed indentation
disabled test_env_semantics

* fix unit tests

* fixes

* CubeCrash, MemorizeDigits tested

* gym backwards compatibility patch

* gym backwards compatibility, followup fixes

* changelist, add spaces to main namespaces

* undo_logger_setup for backwards compat

* remove configuration.py

examples/agents/keyboard_agent.py

@@ -1,10 +1,12 @@
 #!/usr/bin/env python
 from __future__ import print_function
 
-import sys, gym
+import sys, gym, time
 
 #
-# Test yourself as a learning agent! Pass environment name as a command-line argument.
+# Test yourself as a learning agent! Pass environment name as a command-line argument, for example:
+#
+# python keyboard_agent.py SpaceInvadersNoFrameskip-v4
 #
 
 env = gym.make('LunarLander-v2' if len(sys.argv)<2 else sys.argv[1])
@@ -12,7 +14,6 @@ env = gym.make('LunarLander-v2' if len(sys.argv)<2 else sys.argv[1])
 if not hasattr(env.action_space, 'n'):
     raise Exception('Keyboard agent only supports discrete action spaces')
 ACTIONS = env.action_space.n
-ROLLOUT_TIME = 1000
 SKIP_CONTROL = 0    # Use previous control decision SKIP_CONTROL times, that's how you
                     # can test what skip is still usable.
 
@@ -44,26 +45,36 @@ def rollout(env):
     human_wants_restart = False
     obser = env.reset()
     skip = 0
-    for t in range(ROLLOUT_TIME):
+    total_reward = 0
+    total_timesteps = 0
+    while 1:
         if not skip:
             #print("taking action {}".format(human_agent_action))
             a = human_agent_action
+            total_timesteps += 1
             skip = SKIP_CONTROL
         else:
             skip -= 1
 
         obser, r, done, info = env.step(a)
-        env.render()
+        if r != 0:
+            print("reward %0.3f" % r)
+        total_reward += r
+        window_still_open = env.render()
+        if window_still_open==False: return False
         if done: break
         if human_wants_restart: break
         while human_sets_pause:
             env.render()
-            import time
             time.sleep(0.1)
+        time.sleep(0.1)
+    print("timesteps %i reward %0.2f" % (total_timesteps, total_reward))
 
 print("ACTIONS={}".format(ACTIONS))
 print("Press keys 1 2 3 ... to take actions 1 2 3 ...")
 print("No keys pressed is taking action 0")
 
 while 1:
-    rollout(env)
+    window_still_open = rollout(env)
+    if window_still_open==False: break
+