Commit Graph

32 Commits

Author SHA1 Message Date
Jie Tang
28913dad94 Add retries to benchmark upload 2017-03-06 13:46:37 -08:00
Jie Tang
07d9885471 Tighten up this docstring 2017-03-06 13:46:37 -08:00
Jie Tang
f869ed01fe Add skip_videos flag for skipping video uploads on large benchmarks 2017-03-06 13:46:37 -08:00
Jie Tang
ffc30879de Refactor benchmark / evaluation uploads 2017-03-06 13:46:37 -08:00
Tom Brown
d337f4e571 TimeLimit refactor with Monitor Simplification (#482)
* fix double reset, as suggested by @jietang

* better floors and ceilings

* add convenience methods to monitor

* add wrappers to gym namespace

* allow playing Atari games, with potentially more coming in the future

* simplify example in docs

* Move play out of the Env

* fix tests

* no more deprecation warnings

* remove env.monitor

* monitor simplification

* monitor simplifications

* monitor related fixes

* a few changes suggested by linter

* timestep_limit fixes

* keep track of gym env variables for future compatibility

* timestep_limit => max_episode_timesteps

* don't apply TimeLimit wrapper in make for VNC envs

* Respect old timestep_limit argument

* Pass max_episode_seconds through registration

* Don't include deprecation warnings yet
2017-02-01 13:10:59 -08:00
catherio
ce441cd0f8 Logging message formatting typo 2017-01-31 13:58:30 -08:00
Szymon Sidor
a4eb1ac9d4 add ability to submit tags to benchmark_run using gym.upload 2017-01-04 10:41:45 +01:00
Tom Brown
2d44ed4968 Add Monitored wrapper (#434)
* Add WIP Monitored wrapper

* Remove irrelevant render after close monitor test

* py27 compatibility

* Fix test_benchmark

* Move Monitored out of wrappers __init__

* Turn Monitored into a function that returns a Monitor class

* Fix monitor tests

* Remove deprecated test

* Remove deprecated utility

* Prevent duplicate wrapping, add test

* Fix test

* close env in tests to prevent writing to nonexistent file

* Disable semisuper tests

* typo

* Fix failing spec

* Fix monitoring on semisuper tasks

* Allow disabling of duplicate check

* Rename MonitorManager

* Monitored -> Monitor

* Clean up comments

* Remove cruft
2016-12-23 16:21:42 -08:00
Jie Tang
71bb5f8563 Stop seeding in monitor / uploading seeds to scoreboard 2016-10-31 23:56:37 -07:00
Jie Tang
9347e0611b Error if no evaluations found 2016-10-27 20:22:22 -07:00
Jie Tang
7513f6e2bd Allow mismatched uploads 2016-10-25 21:55:54 -07:00
Jie Tang
1f7c6464b7 Thread data_source and initial_reset_timestamps through to scoreboard 2016-10-20 22:19:39 -07:00
Jie Tang
a780b75556 Update upload to respect new gym benchmark spec format 2016-10-20 21:10:34 -07:00
Greg Brockman
517c2e0524 Tweak error message 2016-10-18 20:29:38 -07:00
Jie Tang
db4f977bcd Treat each directory as a single trial 2016-10-18 20:26:08 -07:00
Jie Tang
2dc3e56ac2 Don't check exact seeds for now 2016-10-18 12:05:01 -07:00
Jie Tang
6e0f83e9d5 Validate multiple seeds for benchmark 2016-10-17 17:56:14 -07:00
Jie Tang
422882913b Separate benchmark and evaluation control flow in upload(), add client side validation of benchmark contents 2016-10-09 20:49:49 -07:00
Greg Brockman
934b2acbb7 Add benchmark support (#338)
* Warn if seed doesn't return a list

* Add preliminary BenchmarkRun support

* Add experimental benchmark registration

* Flesh out interface

* Add preliminary BenchmarkRun support

* Warn if seed doesn't return a list

* Add experimental benchmark registration

* Flesh out interface

* Make benchmarkrun upload recursive

* Add evaluation episodes

* Add benchmark scoring

* Tweak reward locations

* Tweak scoring

* Clear default metadata in Wrapper

* Improve scoring

* Expose registry; fix test

* Add initial_reset_timestamp

* Add back algorithm; fix tests
2016-09-23 01:04:26 -07:00
Philip Paquette
f4ae35ea73 Wrappers - Added 'wrappers' and 'step_count' property + monitor support (#288)
* Wrappers - Added 'wrappers' and 'step_count' property + monitor support

* Removed step_count and wrappers from api and monitor

* Removed wrappers and name property
2016-08-13 10:25:19 -07:00
Jie Tang
d173281ea6 Remove algorithm id from example agents, add documentation to api.py 2016-06-01 11:16:29 -07:00
Greg Brockman
58e6aa95e5 [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
2016-05-29 09:07:09 -07:00
rafaelcp
2e26518b4f Fixes tar file generation for upload on Windows (#129)
* Fixes tar file generation for upload on Windows

The temp file cannot be added to the tar while open on Windows.

* Update api.py

* Rewritten with try..finally

* Fix indent.
2016-05-27 23:52:27 -07:00
Jie Tang
b92796ae89 Renaming close_registry to closer 2016-05-18 01:27:58 -07:00
Jie Tang
b578a22bef Refactor monitor to use CloseRegistry 2016-05-16 23:45:56 -07:00
Trevor Blackwell
d845b79a07 Allow resuming monitoring 2016-05-10 17:25:23 -07:00
John Schulman
b69df1b5f3 fix subsampling of videos 2016-05-02 22:43:17 -07:00
Jonas Schneider
7ec98ede9b Cast WeakRefDict to list first to ensure stability 2016-05-01 23:39:57 -04:00
Jonas Schneider
e45f945afd Python 3 encoding fixes
Closes #43.
2016-05-01 23:39:48 -04:00
Greg Brockman
98461ff8d8 Add env_id to log messages while uploading 2016-04-28 07:44:03 -07:00
Jonas Schneider
0a543edc72 gym.scoreboard: Correctly pass a given API key 2016-04-27 12:46:35 -07:00
Greg Brockman
e8f2980603 Initial release. Hello world :). 2016-04-27 08:00:58 -07:00