dodox/metastable-baselines
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity
140 Commits 1 Branch 0 Tags
0e4eedae5e02231d8d781ddd917090b4536a4247
Commit Graph
100 Commits
This Branch
This Branch
All Branches
Author SHA1 Message Date
dodox 0e4eedae5e Fixed gradient throught spherical-chol 2022-08-10 11:55:08 +02:00
dodox 520dc98eb5 Implemented SDE 2022-08-10 11:54:52 +02:00
dodox 12e422aec7 Why does KL double free? 2022-08-07 18:04:40 +02:00
dodox 75d73049b4 Fixing bugs with w2 and sqrt_induced_gaussian 2022-08-06 21:25:49 +02:00
dodox 802094a50f Enabled w2 (can now get sqrt from dist) 2022-08-06 14:54:59 +02:00
dodox 508ebf51f0 Implemented sqrt-induced-gaussian for W2-Projection 2022-08-06 14:46:42 +02:00
dodox fcd9953b37 Testing... 2022-08-06 14:37:30 +02:00
dodox 54113fd40c Removed unused dependency 2022-08-06 14:37:06 +02:00
dodox 2c1689fbbc Fixed smol bugs when instanciating based on string-names of enums 2022-08-06 14:36:35 +02:00
dodox e074294b88 +1 2022-08-05 21:07:38 +02:00
dodox 8b82347056 Automatic casting to enums 2022-08-05 21:06:31 +02:00
dodox 683644f77d Removed weird line... 2022-08-05 21:06:05 +02:00
dodox a78e81b9e1 Allowing prob squashing 2022-07-21 09:42:25 +02:00
dodox 199ce0c8cb ProbSquashing implemented (tanh) 2022-07-20 10:32:19 +02:00
dodox 05dad44b6e Support SAC for testing 2022-07-19 10:08:47 +02:00
dodox 6384d411a9 Support SAC in replays 2022-07-19 10:08:34 +02:00
dodox 0162a36824 Renamed TRL_SAC => SAC 2022-07-19 10:08:14 +02:00
dodox 7b667e9650 SAC is back; with SDC; without Projections 2022-07-19 10:07:50 +02:00
dodox 5f32435751 Smashing bugs: dont confuse chol with chol_net 2022-07-19 10:07:20 +02:00
dodox b7de99b1fc EnforcePositiveType makes no sense for Strength.NONE 2022-07-19 10:06:40 +02:00
dodox 9133ecd61b Show confidence-ellipsoid for supported envs 2022-07-17 00:48:17 +02:00
dodox 49f9acff3e Fixed: Wrong simplification for Hybrid[SCALAR=>FULL] 2022-07-17 00:47:47 +02:00
dodox 046fa78206 Fixed: _chol_from_sphe_chol was unable to handle batches 2022-07-16 17:34:25 +02:00
dodox c141599662 Fixed Bug: _chol_from_sphe_chol dependet on action_dim, not able to use 
for Hybrid methods
 2022-07-16 15:47:09 +02:00
dodox 3fa6de7e66 Broader sampling of stds for logging with batched full covs 2022-07-16 15:28:16 +02:00
dodox bc0e188a0d Removed debugging-code 2022-07-16 15:19:56 +02:00
dodox d2d84d3287 Fixed bug for logging std-estimates when using batched data 2022-07-16 15:18:24 +02:00
dodox 4a24381f46 Fixed bug when using batches with SPHERICAL_CHOL 2022-07-16 15:17:48 +02:00
dodox 04529e8261 Removed debug-point 2022-07-16 14:58:29 +02:00
dodox 4854346f2d Fixed bug with logging of std for full-cov 2022-07-16 14:58:00 +02:00
dodox cb9ee4f302 Fixed bugs for Hybrid[Diag=>Full] 2022-07-16 14:57:34 +02:00
dodox ad584d70fd Removed debugging prints 2022-07-16 13:07:08 +02:00
dodox 72754525cd Allow using newly implemented hybrid method 2022-07-16 13:06:07 +02:00
dodox fa167b3e5f Hybrid Diag -> Full Implemented; Made spherical_chol more efficient 2022-07-16 13:05:35 +02:00
dodox f184b88f19 Allow std logging for full and diagonal cov policies 2022-07-15 18:46:42 +02:00
dodox 74697e8773 Smol bug fixes 2022-07-15 18:46:17 +02:00
dodox 2e0f46b0f3 Fixing ser/deser bug (cloudpickle cant handle some enums) 2022-07-15 18:45:38 +02:00
dodox a86d19053d Smashing bugs (dimension mismatch between Normal and 
Independent/MultivariateNormal)
 2022-07-15 15:46:31 +02:00
dodox ab557a8856 Making MultivariateNormal Policies work (and porting Normal to 
Independent)
 2022-07-15 15:03:51 +02:00
dodox b1ed9fc2b8 Renamed TRL_PG to PPO 2022-07-13 19:51:33 +02:00
dodox 1706bea571 Testing SDC 2022-07-13 19:39:09 +02:00
dodox 3304fd49f6 Working on UniversalGaussianDistribution 2022-07-13 19:38:57 +02:00
dodox fae19509bc Implemented Policies with Contextual Covariance 2022-07-13 19:38:20 +02:00
dodox 41e4170b2f Fixes + spherical_chol 2022-07-11 17:28:08 +02:00
dodox e4440428f8 Working on SDC 2022-07-11 11:55:23 +02:00
dodox 4c4b12ee0e Allow cloning UniversalGaussianDistribution (new_dist_like) 2022-07-09 14:46:11 +02:00
dodox c08ea1cb91 Making UniversalGaussianDistribution ready for tanh-squashing-support 2022-07-09 14:33:07 +02:00
dodox 249754ee89 Wrote a little helper-function to generate all allowed combinations of 
cov-parameterizations
 2022-07-09 14:03:56 +02:00
dodox e09950b30c Work on Contextual Covariances 2022-07-09 12:26:39 +02:00
dodox aacacebfc4 Fixed license 2022-07-06 13:26:33 +02:00
dodox 0911a04e98 Factored out Gaussian Collection for RolloutBuffer 2022-07-04 21:21:16 +02:00
dodox 1f3eac5398 Cleanup 2022-07-02 16:42:14 +02:00
dodox 92204c448f Testing kl-projection (it's working!) 2022-07-02 16:41:25 +02:00
dodox ad7ed0071b Silence! 2022-07-01 20:02:29 +02:00
dodox 91f64c10d7 Fixed bug with initialization of buffer 2022-07-01 20:02:09 +02:00
dodox 0dc9edf112 We no longer use venv (breaks cpp_projection...) 2022-07-01 19:52:22 +02:00
dodox 14100cccc8 Working on SDC-impl. 2022-07-01 15:14:41 +02:00
dodox 4f2e75b7ae Slimmed 3rd-party-licenses 2022-07-01 14:03:50 +02:00
dodox 90b3c68a56 Maybe better licensing 2022-07-01 13:59:07 +02:00
dodox cc51547681 Fixed CppProjection in requirements 2022-07-01 13:47:04 +02:00
dodox 84d1cda96c Trying to get kl to work 2022-07-01 13:45:58 +02:00
dodox 9d7ce73a0b README: We also have to install cpp_projection 2022-07-01 12:30:29 +02:00
dodox 111b1b846d Now we even include a setup.py 2022-07-01 12:22:27 +02:00
dodox 81ae3e3707 Finalized venv support and added installation-instructions 2022-07-01 12:19:57 +02:00
dodox f25d7a103b Now ready to pvenv 2022-07-01 12:03:20 +02:00
dodox 2626ec82a6 Added code-src-note 2022-07-01 11:52:50 +02:00
dodox ab1b269af9 Allow checking whether a dist is contextual 2022-07-01 11:52:14 +02:00
dodox a8b9c63965 Making dez covariances contextual 2022-07-01 11:29:12 +02:00
dodox 155a298e41 Switching to new icon 2022-06-30 21:02:22 +02:00
dodox 7e7ba65e51 Testing alternative icon 2022-06-30 21:01:25 +02:00
dodox 2e378d0a7d Rebranding to Metastable Baselines 2022-06-30 20:40:30 +02:00
dodox 30c9e93967 Fixed replay 2022-06-29 17:02:40 +02:00
dodox 28561b9bb2 Allow manual early stopping of training (Ctrl+C) 2022-06-29 12:46:57 +02:00
dodox e8d423f91f Testing the new WassersteinProjectionLayer 2022-06-29 12:46:37 +02:00
dodox 4e77190d8e Fixed chol not expanding bug and function to shrink chol to diag 2022-06-29 12:44:13 +02:00
dodox 7c117cfca5 Added possibility to load models and run hem again (currently bugged) 2022-06-29 12:43:21 +02:00
dodox 416dde202d Factored out frob_sq and perf improvement for spd input 2022-06-27 13:44:08 +02:00
dodox f4c87c9cdc Better handling of diagonal-covariance as vector and matrix 2022-06-26 18:14:12 +02:00
dodox bc61a6db32 Refactored some stuff out 2022-06-26 16:39:37 +02:00
dodox edf00553dd Renamed our RolloutBuffer and testing the FrobeniusProjectionLayer 2022-06-26 16:39:06 +02:00
dodox 024a9a0265 StillTesting 2022-06-26 16:38:46 +02:00
dodox 80741776d2 Removed old comments 2022-06-25 21:56:07 +02:00
dodox b8488c531b Implemented TRLRolloutBuffer 2022-06-25 21:47:39 +02:00
dodox 60c954c8c1 LunarLanderContinuous-v2 is our new default test-env 2022-06-25 21:47:21 +02:00
dodox cf5a2e82fc mean and std are now saved to the rollout 2022-06-25 18:29:55 +02:00
dodox df21e1dc3f Fixed reference to gym.spaces 2022-06-25 17:49:46 +02:00
dodox 5f7cfd2e10 Note about Code-Src 2022-06-25 15:37:45 +02:00
dodox b9f66dd95d Using BaseProjectionLayer as Default 2022-06-25 15:33:43 +02:00
dodox cafc90409f _global_steps ctr added 2022-06-25 15:05:54 +02:00
dodox 866f863d70 overriding collect_rollouts 2022-06-25 14:57:27 +02:00
dodox 1a49a412c0 expanding automatic testing 2022-06-25 14:50:19 +02:00
dodox 25316ec0b8 Incremental progress at implementing trl_pg 2022-06-25 13:58:58 +02:00
dodox 941b7347f1 Get FPS from env 2022-06-22 13:12:55 +02:00
dodox 0e17b4c07e Fixed model storage location bug 2022-06-22 13:00:40 +02:00
dodox 41d4e94dbe Extended .gitignore 2022-06-22 13:00:28 +02:00
dodox 84b3710850 Testing SACs ability to solve EasierObstacles-v0 2022-06-21 15:15:38 +02:00
dodox 13d335f856 models added to .gitignore 2022-06-21 15:15:22 +02:00
dodox e71735bf79 Trying to converge on simple columbus envs 2022-06-20 23:12:42 +02:00
dodox b9303416ac Split PPO_SDE into PPO_BASE_SDE and PPO_LATENT_SDE 2022-06-19 22:47:04 +02:00
dodox c0a5b6650d Added model*.py to .gitignore 2022-06-19 20:34:33 +02:00