[WIP] add support for seeding environments (#135)

* Make environments seedable

* Fix monitor bugs

- Set monitor_id before setting the infix. This was a bug that would yield incorrect results with multiple monitors.
- Remove extra pid from stats recorder filename. This should be purely cosmetic.

* Start uploading seeds in episode_batch

* Fix _bigint_from_bytes for python3

* Set seed explicitly in random_agent

* Pass through seed argument

* Also pass through random state to spaces

* Pass random state into the observation/action spaces

* Make all _seed methods return the list of used seeds

* Switch over to np.random where possible

* Start hashing seeds, and also seed doom engine

* Fixup seeding determinism in many cases

* Seed before loading the ROM

* Make seeding more Python3 friendly

* Make the MuJoCo skipping a bit more forgiving

* Remove debugging PDB calls

* Make setInt argument into raw bytes

* Validate and upload seeds

* Skip box2d

* Make seeds smaller, and change representation of seeds in upload

* Handle long seeds

* Fix RandomAgent example to be deterministic

* Handle integer types correctly in Python2 and Python3

* Try caching pip

* Try adding swap

* Add df and free calls

* Bump swap

* Bump swap size

* Try setting overcommit

* Try other sysctls

* Try fixing overcommit

* Try just setting overcommit_memory=1

* Add explanatory comment

* Add what's new section to readme

* BUG: Mark ElevatorAction-ram-v0 as non-deterministic for now

* Document seed

* Move nondetermistic check into spec

gym/envs/doom/doom_basic.py

@@ -6,6 +6,7 @@ import numpy as np
 from doom_py import DoomGame, Mode, Button, GameVariable, ScreenFormat, ScreenResolution, Loader
 from gym import error, spaces
 from gym.envs.doom import doom_env
+from gym.utils import seeding
 
 logger = logging.getLogger(__name__)
 
@@ -49,10 +50,19 @@ class DoomBasicEnv(doom_env.DoomEnv):
         self.game.set_doom_map('map01')
         self.screen_height = 480                    # Must match .cfg file
         self.screen_width = 640                     # Must match .cfg file
-        # 3 allowed actions [0, 9, 10] (must match .cfg file)
-        self.action_space = spaces.HighLow(np.matrix([[0, 1, 0]] * 3))
-        self.observation_space = spaces.Box(low=0, high=255, shape=(self.screen_height, self.screen_width, 3))
         self.game.set_window_visible(False)
         self.viewer = None
         self.game.init()
         self.game.new_episode()
+        self._seed()
+
+    def _seed(self, seed=None):
+        np_random, seed1 = seeding.np_random(seed)
+        # Derive a random seed.
+        seed2 = seeding.hash_seed(seed1 + 1) % 2**32
+        self.game.set_seed(seed2)
+
+        # 3 allowed actions [0, 9, 10] (must match .cfg file)
+        self.action_space = spaces.HighLow(np.matrix([[0, 1, 0]] * 3), np_random=np_random)
+        self.observation_space = spaces.Box(low=0, high=255, shape=(self.screen_height, self.screen_width, 3), np_random=np_random)
+        return [seed1, seed2]