Peter Zhokhov
1c5c6563b7
reverted VecNormalize to use RunningMeanStd (no tf)
2018-08-02 10:55:09 -07:00
Peter Zhokhov
f6d1115ead
reverted running_mean_std to user property decorators for mean, var, count
2018-08-02 10:32:22 -07:00
Peter Zhokhov
f6d5a47bed
use ncpu=1 for mujoco sessions - gives a bit of a performance speedup
2018-08-02 10:24:21 -07:00
Peter Zhokhov
c2df27bee4
non-tf normalization benchmark RUN BENCHMARKS
2018-08-02 09:41:41 -07:00
Peter Zhokhov
974c15756e
changed default ppo2 lr schedule to linear RUN BENCHMARKS
2018-08-01 16:24:44 -07:00
Peter Zhokhov
ad43fd9a35
add defaults
2018-08-01 16:15:59 -07:00
Peter Zhokhov
72c357c638
hardcode names of retro environments
2018-08-01 15:18:59 -07:00
Peter Zhokhov
e00e5ca016
run ppo_mpi benchmarks only RUN BENCHMARKS
2018-08-01 14:56:08 -07:00
Peter Zhokhov
705797f2f0
Merge branch 'peterz_migrate_rlalgs' into peterz_benchmarks
2018-08-01 14:46:40 -07:00
Peter Zhokhov
fcd84aa831
make_atari_env compatible with mpi
2018-08-01 14:46:18 -07:00
Peter Zhokhov
390b51597a
benchmarks on ppo2 only RUN BENCHMARKS
2018-08-01 11:01:50 -07:00
Peter Zhokhov
95104a3592
Merge branch 'peterz_migrate_rlalgs' into peterz_benchmarks
2018-08-01 10:50:29 -07:00
Peter Zhokhov
3528f7b992
save all variables to make sure we save the vec_normalize normalization
2018-08-01 10:12:19 -07:00
Peter Zhokhov
151e48009e
flake8 complaints
2018-07-31 16:25:12 -07:00
Peter Zhokhov
92f33335e9
dummy commit to RUN BENCHMARKS
2018-07-31 15:53:18 -07:00
Peter Zhokhov
af729cff15
dummy commit to RUN BENCHMARKS
2018-07-31 15:37:00 -07:00
Peter Zhokhov
10f815fe1d
fixed import in vec_normalize
2018-07-31 15:19:43 -07:00
Peter Zhokhov
8c4adac898
running_mean_std uses tensorflow variables
2018-07-31 14:45:55 -07:00
Peter Zhokhov
2a93ea8782
serialize variables as a dict, not as a list
2018-07-31 11:13:31 -07:00
Peter Zhokhov
9c48f9fad5
very dummy commit to RUN BENCHMARKS
2018-07-31 10:23:43 -07:00
Peter Zhokhov
348cbb4b71
dummy commit to RUN BENCHMARKS
2018-07-31 09:42:23 -07:00
Peter Zhokhov
a1602ab15f
dummy commit to RUN BENCHMARKS
2018-07-30 17:51:16 -07:00
Peter Zhokhov
e63e69bb14
dummy commit to RUN BENCHMARKS
2018-07-30 17:39:22 -07:00
Peter Zhokhov
385e7e5c0d
dummy commit to RUN BENCHMARKS
2018-07-30 17:21:05 -07:00
Peter Zhokhov
d112a2e49f
added approximate humanoid reward with ppo2 into the README for reference
2018-07-30 16:58:31 -07:00
Peter Zhokhov
e662dd6409
run.py can run algos from both baselines and rl_algs
2018-07-30 16:09:48 -07:00
Peter Zhokhov
efc6bffce3
replaced atari_arg_parser with common_arg_parser
2018-07-30 15:58:56 -07:00
Peter Zhokhov
872181d4c3
re-exported rl_algs - fixed problems with serialization test and test_cartpole
2018-07-30 15:49:48 -07:00
Peter Zhokhov
628ddecf6a
re-exported rl_algs
2018-07-30 12:15:46 -07:00
peter
83a4a4be65
run slow tests
2018-07-26 14:39:25 -07:00
peter
7edac38c73
more stuff from rl-algs
2018-07-26 14:26:57 -07:00
peter
a6dca44115
exported rl-algs
2018-07-26 14:02:04 -07:00
Adam Gleave
f272969325
GAIL: bugfix in dataset loading ( #447 )
...
* Fix silly typo
* Replace ad-hoc function with NumPy code
2018-07-06 16:12:14 -07:00
pzhokhov
a6b1bc70f1
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
2018-06-08 09:41:45 -07:00
pzhokhov
36ee5d1707
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
2018-06-06 11:39:13 -07:00
pzhokhov
24fe3d6576
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
2018-05-21 15:24:00 -07:00
pzhokhov
9cb7ece338
add opencv-python to the dependencies ( #407 )
2018-05-14 10:52:19 -07:00
pzhokhov
9cf95a0054
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
2018-05-03 09:43:28 -07:00
pzhokhov
8b781038cc
put filters and running_stat files in common instead of acktr ( #389 )
2018-05-02 18:42:48 -07:00
pzhokhov
69f25c6028
import internal repo ( #385 )
2018-05-01 16:54:04 -07:00
pzhokhov
2b0283b9db
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
2018-04-25 17:40:48 -07:00
Matthias Plappert
1f8a03f3a6
Update README
2018-03-26 16:50:22 +02:00
Matthias Plappert
3cc7df0608
Minor fixes to HER release ( #319 )
...
* Fix plotting script
* Add warning if num_cpu = 1
2018-03-05 11:06:17 +01:00
Alex Nichol
8b3a6c2051
fix DummyVecEnv reusing buffers
2018-03-02 17:18:07 -08:00
Alex Nichol
569bd42629
Merge pull request #308 from araffin/master
...
Bug fix in saving ACER model
2018-03-01 10:45:04 -08:00
Daniel Ziegler
f49a9c3d85
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.
2018-03-01 18:00:34 +01:00
Antonin RAFFIN
14f2f9328c
Bug fix in saving ACER model
2018-03-01 10:24:14 +01:00
Alex Nichol
6bdf2f55a2
Merge pull request #132 from bhatiaabhinav/bug_fixes
...
Bug fix in saving a2c model.
2018-02-27 19:00:37 -08:00
Alex Nichol
97be70d6c8
fixes for DummyVecEnv
...
Fixes various problems running MuJoCo tasks.
2018-02-27 18:55:10 -08:00
Matthias Plappert
b71152eea0
Adds support for Hindsight Experience Replay (HER) ( #299 )
...
* Add Hindsight Experience Replay (HER)
* Minor improvements
2018-02-26 17:40:16 +01:00
