bcde04e | Youngjin Kim | 16 August 2018, 21:55:57 UTC | Fix argument error in deepq (#508) * Fix argment error in deepq * Fix argment error in deepq | 16 August 2018, 21:55:57 UTC |
5622a09 | pzhokhov | 16 August 2018, 21:53:49 UTC | update readmes (#514) * update per-algorithm READMEs to reflect new way of running algorithms * adding a link to repo-wide README * updated README files and deepq.train_cartpole example | 16 August 2018, 21:53:49 UTC |
e2da7cd | Pim de Haan | 16 August 2018, 19:08:53 UTC | Several bugfixes for #504, #505, #506 related to Classic Control and deepq (#507) * Several bugfixes * Fixed ActWrapper.step bug | 16 August 2018, 19:08:53 UTC |
b222dd0 | Peter Zhokhov | 13 August 2018, 23:01:24 UTC | updated links in README to point to master | 13 August 2018, 23:01:24 UTC |
1870685 | pzhokhov | 13 August 2018, 22:59:43 UTC | Publish benchmark results (#502) * updated benchmark pages with final rewards * use htmlpreview to render pages * use htmlpreview to render pages * use htmlpreview to render pages * updated README to reflect ppo1 being obsolete * removed navbars from published benchmark pages * fixed link in README | 13 August 2018, 22:59:43 UTC |
8c2aea2 | pzhokhov | 13 August 2018, 16:56:44 UTC | refactor a2c, acer, acktr, ppo2, deepq, and trpo_mpi (#490) * exported rl-algs * more stuff from rl-algs * run slow tests * re-exported rl_algs * re-exported rl_algs - fixed problems with serialization test and test_cartpole * replaced atari_arg_parser with common_arg_parser * run.py can run algos from both baselines and rl_algs * added approximate humanoid reward with ppo2 into the README for reference * dummy commit to RUN BENCHMARKS * dummy commit to RUN BENCHMARKS * dummy commit to RUN BENCHMARKS * dummy commit to RUN BENCHMARKS * very dummy commit to RUN BENCHMARKS * serialize variables as a dict, not as a list * running_mean_std uses tensorflow variables * fixed import in vec_normalize * dummy commit to RUN BENCHMARKS * dummy commit to RUN BENCHMARKS * flake8 complaints * save all variables to make sure we save the vec_normalize normalization * benchmarks on ppo2 only RUN BENCHMARKS * make_atari_env compatible with mpi * run ppo_mpi benchmarks only RUN BENCHMARKS * hardcode names of retro environments * add defaults * changed default ppo2 lr schedule to linear RUN BENCHMARKS * non-tf normalization benchmark RUN BENCHMARKS * use ncpu=1 for mujoco sessions - gives a bit of a performance speedup * reverted running_mean_std to user property decorators for mean, var, count * reverted VecNormalize to use RunningMeanStd (no tf) * reverted VecNormalize to use RunningMeanStd (no tf) * profiling wip * use VecNormalize with regular RunningMeanStd * added acer runner (missing import) * flake8 complaints * added a note in README about TfRunningMeanStd and serialization of VecNormalize * dummy commit to RUN BENCHMARKS * merged benchmarks branch | 13 August 2018, 16:56:44 UTC |
366f486 | Tony Yu Cao | 08 August 2018, 17:42:10 UTC | Update README.md (#416) Update Atari example | 08 August 2018, 17:42:10 UTC |
f272969 | Adam Gleave | 06 July 2018, 23:12:14 UTC | GAIL: bugfix in dataset loading (#447) * Fix silly typo * Replace ad-hoc function with NumPy code | 06 July 2018, 23:12:14 UTC |
a6b1bc7 | pzhokhov | 08 June 2018, 16:41:45 UTC | re-import internal; fix missing tile_images.py (#427) * import rl-algs from 2e3a166 commit * extra import of the baselines badge * exported commit with identity test * proper rng seeding in the test_identity * import internal * adding missing tile_images.py | 08 June 2018, 16:41:45 UTC |
36ee5d1 | pzhokhov | 06 June 2018, 18:39:13 UTC | Import internal changes (#422) * import rl-algs from 2e3a166 commit * extra import of the baselines badge * exported commit with identity test * proper rng seeding in the test_identity * import internal | 06 June 2018, 18:39:13 UTC |
24fe3d6 | pzhokhov | 21 May 2018, 22:24:00 UTC | Import internal repo (#409) * import rl-algs from 2e3a166 commit * extra import of the baselines badge * exported commit with identity test * proper rng seeding in the test_identity | 21 May 2018, 22:24:00 UTC |
9cb7ece | pzhokhov | 14 May 2018, 17:52:19 UTC | add opencv-python to the dependencies (#407) | 14 May 2018, 17:52:19 UTC |
9cf95a0 | pzhokhov | 03 May 2018, 16:43:28 UTC | setup travis ci build (#388) * simple .travis.yml file * added static syntax checks of common to .travis.yml * dockerizing the build * fix Dockerfile, adding build shield * cleaning up workdir in Dockerfile and .travis.yml * .travis.yml fixed common -> baselines/common for style check | 03 May 2018, 16:43:28 UTC |
8b78103 | pzhokhov | 03 May 2018, 01:42:48 UTC | put filters and running_stat files in common instead of acktr (#389) | 03 May 2018, 01:42:48 UTC |
69f25c6 | pzhokhov | 01 May 2018, 23:54:04 UTC | import internal repo (#385) | 01 May 2018, 23:54:04 UTC |
2b0283b | pzhokhov | 26 April 2018, 00:40:48 UTC | Readme.md detailed installation instructions (#377) * changes to README.md files with more detailed installation instructions * md-fying the changes better * link on the word homebrew in readme.md * typos in README.md * README.md * removed extra comma sign * removed sudo from brew command | 26 April 2018, 00:40:48 UTC |
1f8a03f | Matthias Plappert | 26 March 2018, 14:50:22 UTC | Update README | 26 March 2018, 14:50:22 UTC |
3cc7df0 | Matthias Plappert | 05 March 2018, 10:06:17 UTC | Minor fixes to HER release (#319) * Fix plotting script * Add warning if num_cpu = 1 | 05 March 2018, 10:06:17 UTC |
8b3a6c2 | Alex Nichol | 03 March 2018, 01:18:07 UTC | fix DummyVecEnv reusing buffers | 03 March 2018, 01:18:07 UTC |
569bd42 | Alex Nichol | 01 March 2018, 18:45:04 UTC | Merge pull request #308 from araffin/master Bug fix in saving ACER model | 01 March 2018, 18:45:04 UTC |
f49a9c3 | Daniel Ziegler | 01 March 2018, 17:00:34 UTC | Fix bug in DDPG parameter space noise adaptation (#306) The training loop used the rollout step variable `t` rather than the training step variable `t_train` to decide when to adapt the scale of the parameter space noise. | 01 March 2018, 17:00:34 UTC |
14f2f93 | Antonin RAFFIN | 01 March 2018, 09:24:14 UTC | Bug fix in saving ACER model | 01 March 2018, 09:24:14 UTC |
6bdf2f5 | Alex Nichol | 28 February 2018, 03:00:37 UTC | Merge pull request #132 from bhatiaabhinav/bug_fixes Bug fix in saving a2c model. | 28 February 2018, 03:00:37 UTC |
97be70d | Alex Nichol | 28 February 2018, 02:55:10 UTC | fixes for DummyVecEnv Fixes various problems running MuJoCo tasks. | 28 February 2018, 02:55:10 UTC |
b71152e | Matthias Plappert | 26 February 2018, 16:40:16 UTC | Adds support for Hindsight Experience Replay (HER) (#299) * Add Hindsight Experience Replay (HER) * Minor improvements | 26 February 2018, 16:40:16 UTC |
df2e846 | Christopher Hesse | 15 February 2018, 06:01:16 UTC | export: fix accidental rename | 15 February 2018, 06:01:16 UTC |
edb52c2 | Christopher Hesse | 15 February 2018, 05:42:22 UTC | export: Fix deepq param noise refactoring, remove atari experiments and azure dependency | 15 February 2018, 05:42:22 UTC |
98257ef | Andrei Kashin | 02 October 2017, 22:29:07 UTC | Flush temporary file before compressing it. We need to flush the buffer after `pickle.dump`, otherwise the resulting zip archive might be incomplete (reproducible, if the state consists of a single integer). | 06 February 2018, 15:04:44 UTC |
d9b3660 | Oleg Klimov | 05 February 2018, 20:25:05 UTC | comment about loading weights in ppo2 | 05 February 2018, 20:25:05 UTC |
2793971 | Oleg Klimov | 05 February 2018, 15:51:27 UTC | fix gail tf_util usage | 05 February 2018, 15:51:27 UTC |
16d7d23 | John Schulman | 03 February 2018, 07:04:53 UTC | Merge pull request #271 from simontudo/add-requirement-cloudpickle added cloudpickle to requirements | 03 February 2018, 07:04:53 UTC |
9175b77 | John Schulman | 03 February 2018, 07:03:51 UTC | Merge pull request #273 from simontudo/videorecorder-import updated videorecorder import | 03 February 2018, 07:03:51 UTC |
615870a | simontudo | 01 February 2018, 11:09:08 UTC | updated videorecorder import | 01 February 2018, 11:09:08 UTC |
7bd264e | simontudo | 31 January 2018, 09:43:17 UTC | added cloudpickle to requirements | 31 January 2018, 09:43:17 UTC |
8d03102 | John Schulman | 29 January 2018, 08:54:51 UTC | Merge pull request #265 from 20chase/patch-1 fix logger error for trpo_mpi | 29 January 2018, 08:54:51 UTC |
4a77855 | 20chase | 29 January 2018, 08:52:01 UTC | using mujoco_arg_parser as args remove origin parser | 29 January 2018, 08:52:01 UTC |
2e29b41 | John Schulman | 28 January 2018, 02:42:31 UTC | Merge pull request #268 from ei-grad/master Fix fc call in AcerLstmPolicy | 28 January 2018, 02:42:31 UTC |
634e37c | Andrew Grigorev | 27 January 2018, 20:18:02 UTC | Fix fc call in AcerLstmPolicy The `act` keyword was removed from baselines.a2c.utils.fc in commit 9fa8e1b. | 27 January 2018, 20:18:02 UTC |
452b548 | 20chase | 26 January 2018, 06:34:01 UTC | Merge branch 'master' into patch-1 | 26 January 2018, 06:34:01 UTC |
ebb8aff | John Schulman | 26 January 2018, 05:17:40 UTC | fix trpo_mpi bug where logstd wasn’t included | 26 January 2018, 05:17:40 UTC |
c9613b2 | John Schulman | 26 January 2018, 04:54:34 UTC | Merge pull request #259 from andrewliao11/openai_gail Add gail maintainer list | 26 January 2018, 04:54:34 UTC |
459f007 | John Schulman | 26 January 2018, 04:54:20 UTC | Merge pull request #260 from uidilr/master Add GAIL | 26 January 2018, 04:54:20 UTC |
9fa8e1b | John Schulman | 26 January 2018, 02:33:48 UTC | Lots of cleanups Fixes for new gym version Add @olegklimov and @unixpickle to authors list | 26 January 2018, 02:54:24 UTC |
ac2ea4f | 20chase | 25 January 2018, 14:09:00 UTC | fix logger error for MPI Can't run logger.configure() if rank != 0 | 25 January 2018, 14:09:00 UTC |
d8cce23 | Yusuke Nakata | 23 January 2018, 03:00:00 UTC | Add GAIL | 23 January 2018, 03:02:03 UTC |
0c207f0 | andrew | 22 January 2018, 06:13:01 UTC | fix typo | 22 January 2018, 06:13:01 UTC |
41d41fa | andrew | 22 January 2018, 06:12:03 UTC | add gail maintainer list | 22 January 2018, 06:12:03 UTC |
b5be53d | John Schulman | 22 January 2018, 01:30:20 UTC | Merge pull request #229 from andrewliao11/gail GAIL implementation | 22 January 2018, 01:30:20 UTC |
49c1a8e | Matthias Plappert | 16 January 2018, 18:24:30 UTC | Fix bug in parameter space noise DQN | 16 January 2018, 18:24:30 UTC |
e5a714b | andrew | 12 January 2018, 23:12:45 UTC | fix relative import | 12 January 2018, 23:12:45 UTC |
f9d1d33 | John Schulman | 12 January 2018, 19:05:29 UTC | remove mpirun from ppo2 instructions | 12 January 2018, 19:05:29 UTC |
8c90f67 | Alex Nichol | 15 December 2017, 23:54:43 UTC | don't list TensorFlow as a requirement fixes #146 A better (more involved) solution might be to check for a TensorFlow installation manually in setup.py and deal with that accordingly. | 15 December 2017, 23:54:43 UTC |
f22bee0 | Andrew | 13 December 2017, 03:03:42 UTC | Add files via upload | 13 December 2017, 03:03:42 UTC |
4acc71f | andrew | 13 December 2017, 02:58:57 UTC | add x, y, axis name | 13 December 2017, 02:58:57 UTC |
2f1b629 | andrew | 13 December 2017, 02:56:00 UTC | Merge branch 'gail' of https://github.com/andrewliao11/baselines into gail | 13 December 2017, 02:56:00 UTC |
00573cf | andrew | 13 December 2017, 02:54:03 UTC | add x, y axis name | 13 December 2017, 02:54:03 UTC |
cfa1236 | Andrew | 12 December 2017, 05:21:56 UTC | Update README.md | 12 December 2017, 05:21:56 UTC |
64288f9 | Andrew | 12 December 2017, 05:19:47 UTC | Update gail-result.md | 12 December 2017, 05:19:47 UTC |
5f647d4 | Andrew | 12 December 2017, 05:18:05 UTC | Update README.md | 12 December 2017, 05:18:05 UTC |
6723455 | Andrew | 12 December 2017, 05:15:30 UTC | Update gail-result.md | 12 December 2017, 05:15:30 UTC |
45a93cf | Andrew | 12 December 2017, 05:06:04 UTC | add training curve from tensorboard | 12 December 2017, 05:06:04 UTC |
11604f7 | andrew | 07 December 2017, 20:08:20 UTC | add download link to readme and add description to python file | 07 December 2017, 20:08:20 UTC |
2444034 | John Schulman | 04 December 2017, 22:07:01 UTC | Merge pull request #194 from ryanjulian/env_lines Force shebang lines to Python 3 | 04 December 2017, 22:07:01 UTC |
041b6b7 | John Schulman | 04 December 2017, 22:02:49 UTC | Merge pull request #215 from chris-chris/feature/typo-2017-11-19 fix misspellings | 04 December 2017, 22:02:49 UTC |
5d62b5b | John Schulman | 04 December 2017, 22:01:38 UTC | Merge pull request #221 from jvmancuso/patch-1 Docstring fix | 04 December 2017, 22:01:38 UTC |
2fcc9b9 | John Schulman | 04 December 2017, 22:01:12 UTC | Merge pull request #226 from definitelyuncertain/master Call ppo2 and not ppo1 in ppo2 README.md | 04 December 2017, 22:01:12 UTC |
0000339 | Andrew | 03 December 2017, 23:50:24 UTC | Update gail-result.md | 03 December 2017, 23:50:24 UTC |
6090ee8 | andrew | 03 December 2017, 23:46:52 UTC | add comparison for expert/BC/gail | 03 December 2017, 23:46:52 UTC |
7954327 | andrew | 03 December 2017, 21:55:44 UTC | add behavior cloning learn/eval code | 03 December 2017, 21:55:44 UTC |
8495890 | andrew | 03 December 2017, 09:49:42 UTC | add gail, file_writer for tf.summary, and allow specifying var_list for tf.train.Saver | 03 December 2017, 09:49:42 UTC |
6431849 | definitelyuncertain | 02 December 2017, 16:30:28 UTC | Call ppo2 and not ppo1 | 02 December 2017, 16:30:28 UTC |
36e074d | jvmancuso | 27 November 2017, 19:45:50 UTC | Update replay_buffer.py | 27 November 2017, 19:45:50 UTC |
c336409 | Ubuntu | 19 November 2017, 01:29:30 UTC | fix misspellings | 19 November 2017, 01:29:30 UTC |
b05be68 | John Schulman | 17 November 2017, 06:14:30 UTC | add missing files, fix Issue #209 | 17 November 2017, 06:14:30 UTC |
2dd7d30 | John Schulman | 16 November 2017, 18:02:32 UTC | Add ACER, PPO2, and results_plotter.py | 16 November 2017, 18:02:32 UTC |
df889ca | Ryan Julian | 05 November 2017, 23:22:16 UTC | Force shebang lines to Python 3 This is a Python 3-only library. A shebang with `#!/usr/bin/env python` will launch python2 on many systems which do not have python3 installed. Setting the shebang to `#!/usr/bin/env python3` will show a useful error on systems without Python 3. | 05 November 2017, 23:22:16 UTC |
6a3cbb4 | John Schulman | 26 October 2017, 02:20:30 UTC | switch append mode to write mode | 26 October 2017, 02:20:30 UTC |
bb40378 | John Schulman | 25 October 2017, 13:21:29 UTC | change atari preprocessing to use faster opencv some logger changes | 25 October 2017, 13:21:29 UTC |
4993286 | John Schulman | 09 October 2017, 21:12:28 UTC | Merge pull request #160 from mkarutz/fixFrameStackingA2C Fixes frame stacking in A2C and ACKTR for multi-channel observations | 09 October 2017, 21:12:28 UTC |
cc8818f | Malcolm Karutz | 09 October 2017, 02:08:41 UTC | Fixes frame stacking in A2C and ACKTR for multi-channel observation spaces. | 09 October 2017, 02:08:41 UTC |
3eb71a0 | John Schulman | 30 September 2017, 21:51:56 UTC | Merge pull request #151 from emansim/master Fixes the NaN issues in ACKTR + bug in run_mujoco.py | 30 September 2017, 21:51:56 UTC |
f8663ea | Elman Mansimov | 30 September 2017, 21:21:04 UTC | fixes acktr_cont issues | 30 September 2017, 21:21:04 UTC |
3d1e171 | Abhinav Bhatia | 11 September 2017, 18:32:31 UTC | Bug fix in saving a2c model. | 11 September 2017, 18:35:43 UTC |
699919f | John Schulman | 07 September 2017, 08:57:04 UTC | Merge pull request #64 from jhumplik/master Use standardized advantages in trpo. | 07 September 2017, 08:57:04 UTC |
498b4cf | John Schulman | 06 September 2017, 08:04:47 UTC | Merge pull request #128 from louiehelm/louiehelm-patch-1 Fix command lines | 06 September 2017, 08:04:47 UTC |
5893874 | Louie Helm | 05 September 2017, 13:06:19 UTC | fix ppo command in readme | 05 September 2017, 13:06:19 UTC |
3d3ea6c | Louie Helm | 05 September 2017, 13:04:37 UTC | fix trpo command in readme | 05 September 2017, 13:04:37 UTC |
902ffcb | John Schulman | 29 August 2017, 04:27:23 UTC | Merge pull request #120 from hamzamerzic/tensorflow_global_variable Deprecated VARIABLES -> GLOBAL_VARIABLES. | 29 August 2017, 04:27:23 UTC |
a7320b8 | Hamza Merzic | 28 August 2017, 14:51:48 UTC | Deprecated VARIABLES -> GLOBAL_VARIABLES. | 28 August 2017, 14:51:48 UTC |
4e2a570 | John Schulman | 28 August 2017, 05:54:52 UTC | Merge pull request #104 from stevenschmatz/patch-1 Fix relative links in README.md | 28 August 2017, 05:54:52 UTC |
6f39148 | John Schulman | 28 August 2017, 05:49:50 UTC | fix gym req | 28 August 2017, 05:49:50 UTC |
2f30833 | John Schulman | 28 August 2017, 05:36:44 UTC | Merge branch 'master' of github.com:openai/baselines | 28 August 2017, 05:36:44 UTC |
00cdeff | John Schulman | 28 August 2017, 05:36:24 UTC | add __init__.py | 28 August 2017, 05:36:24 UTC |
410ef38 | John Schulman | 28 August 2017, 05:31:46 UTC | Merge pull request #103 from learnercys/master Adding links to source files | 28 August 2017, 05:31:46 UTC |
aa6e58b | John Schulman | 28 August 2017, 05:22:14 UTC | fix readmes | 28 August 2017, 05:22:14 UTC |
d9f194f | John Schulman | 28 August 2017, 05:13:48 UTC | Fix atari wrapper (affecting a2c perf) and pposgd mujoco performance - removed vf clipping in pposgd - that was severely degrading performance on mujoco because it didn’t account for scale of returns - switched adam epsilon in pposgd_simple - brought back no-ops in atari wrapper (oops) - added readmes - revamped run_X_benchmark scripts to have standard form - cleaned up DDPG a little, removed deprecated SimpleMonitor and non-idiomatic usage of logger | 28 August 2017, 05:14:59 UTC |
06b071c | Steven Schmatz | 18 August 2017, 17:35:22 UTC | Fix relative links in README.md | 18 August 2017, 17:35:22 UTC |
3f676f7 | John Schulman | 18 August 2017, 16:25:39 UTC | ACKTR + A2C | 18 August 2017, 16:25:39 UTC |
b7966b3 | Carlos Hernandez | 18 August 2017, 06:58:25 UTC | Adding links to source files | 18 August 2017, 07:16:00 UTC |
8822518 | Matthias Plappert | 27 July 2017, 15:10:59 UTC | Parameter space noise for DQN and DDPG (#75) * Export param noise * Update documentation * Final finishing touches | 27 July 2017, 15:10:59 UTC |