[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

examples/agents/random_agent.py

@@ -19,14 +19,18 @@ if __name__ == '__main__':
     logger.setLevel(logging.INFO)
 
     env = gym.make('CartPole-v0' if len(sys.argv)<2 else sys.argv[1])
-    agent = RandomAgent(env.action_space)
 
     # You provide the directory to write to (can be an existing
     # directory, including one with existing data -- all monitor files
     # will be namespaced). You can also dump to a tempdir if you'd
     # like: tempfile.mkdtemp().
     outdir = '/tmp/random-agent-results'
-    env.monitor.start(outdir, force=True)
+    env.monitor.start(outdir, force=True, seed=0)
+
+    # This declaration must go *after* the monitor call, since the
+    # monitor's seeding creates a new action_space instance with the
+    # appropriate pseudorandom number generator.
+    agent = RandomAgent(env.action_space)
 
     episode_count = 100
     max_steps = 200