 dodox
|
9cd25ee484
|
Support passing trajectories into predictiion functions (for pca)
|
2023-05-21 20:16:25 +02:00 |
|
|
dodox
|
d4003e1a68
|
get_past_trajectories will now return a th.Tensor
|
2023-05-21 17:29:03 +02:00 |
|
|
dodox
|
ea7634ca42
|
Revert "typo: trajectory -> traj"
This reverts commit becb8d36e1.
|
2023-05-21 17:26:13 +02:00 |
|
|
dodox
|
becb8d36e1
|
typo: trajectory -> traj
|
2023-05-21 17:22:59 +02:00 |
|
|
dodox
|
69fdd7b103
|
Fix Bug: Let's let thge pca-lib handle this exception
|
2023-05-21 17:19:39 +02:00 |
|
|
dodox
|
c473990fbd
|
Fix smol Bug
|
2023-05-21 17:15:46 +02:00 |
|
|
dodox
|
834b5c9a65
|
Fixed Bug: Can not get past trajectory from rolloutBuffer
|
2023-05-21 17:11:32 +02:00 |
|
|
dodox
|
6a9d7012c7
|
Bugs
|
2023-05-21 17:03:22 +02:00 |
|
|
dodox
|
43418a5e53
|
Tiny Bugfix
|
2023-05-21 16:40:13 +02:00 |
|
|
dodox
|
a1df15f7bc
|
Fixed PCA bug
|
2023-05-21 16:32:27 +02:00 |
|
|
dodox
|
ea4b9851d8
|
Fixed: Binding to PCA used old package name
|
2023-05-21 16:23:51 +02:00 |
|
|
dodox
|
8921421732
|
Fixed Typo
|
2023-05-21 16:14:34 +02:00 |
|
|
dodox
|
2efa6c18fb
|
Experimental Support for PCA
|
2023-05-21 14:27:09 +02:00 |
|
|
dodox
|
76ea3a6326
|
Implemented prior conditioned annealing (untested)
|
2023-04-25 17:05:34 +02:00 |
|
|
dodox
|
09159774d9
|
Fix backward pass error for hybrid-method
|
2023-03-13 20:44:09 +01:00 |
|
|
dodox
|
75feefbe5a
|
Fix: Typos for sde/rex hybrid
|
2023-02-18 12:42:07 +01:00 |
|
|
dodox
|
c62723bef6
|
First implementation of hybrid (bewteen sde and rex)
|
2023-02-18 12:18:46 +01:00 |
|
|
dodox
|
ae9a95cbfc
|
Fix: Typo in SAC
|
2023-02-02 20:00:16 +01:00 |
|
|
dodox
|
71670782b6
|
.
|
2023-01-28 21:47:19 +01:00 |
|
|
dodox
|
b86498fa9a
|
Now softmax sde latent is actually working...
|
2023-01-28 19:48:43 +01:00 |
|
|
dodox
|
e985d892ca
|
New SDE feature: softmax activation of latent
|
2023-01-28 19:42:32 +01:00 |
|
|
dodox
|
ffbf2b3fe5
|
Allow reduced latent sde dim
|
2023-01-27 13:34:28 +01:00 |
|
|
dodox
|
f421dc2ab5
|
Removed hack again...
|
2023-01-26 18:18:38 +01:00 |
|
|
dodox
|
f86c6c736f
|
Hack is back
|
2023-01-26 18:15:06 +01:00 |
|
|
dodox
|
20646f07f3
|
Reverting last two commits (was ugly hack)
|
2023-01-26 17:43:57 +01:00 |
|
|
dodox
|
6f1837bda5
|
calc episodic infos for some fancy envs
|
2023-01-26 17:27:34 +01:00 |
|
|
dodox
|
f37c8caaa4
|
Force include reward in env infos (for vec env)
|
2023-01-26 12:00:18 +01:00 |
|
|
dodox
|
43cc749809
|
Fix Noise Reset Bug for SAC
|
2023-01-15 13:28:40 +01:00 |
|
|
dodox
|
844cdb1319
|
Fixed: Upgrade changing interfaces
|
2023-01-06 18:46:38 +01:00 |
|
|
dodox
|
f3e03916c8
|
Upgrading to SB3 1.7 (probably broke some stuff...)
|
2022-12-13 19:14:28 +01:00 |
|
|
dodox
|
c47d5741ca
|
New Version of SB3
|
2022-12-13 18:38:33 +01:00 |
|
|
dodox
|
9cb7e89b1d
|
MIT License
|
2022-11-13 20:07:08 +01:00 |
|
|
dodox
|
4ec5c65cf2
|
Tiny fix for other envs
|
2022-11-07 13:23:55 +01:00 |
|
|
dodox
|
302dbf6dde
|
Pruning of README 2nd try
|
2022-11-07 12:59:12 +01:00 |
|
|
dodox
|
c5ed794325
|
Updated requirements
|
2022-11-07 12:55:15 +01:00 |
|
|
dodox
|
d89609e276
|
Merge branch 'master' of dominik-roth.eu:dodox/metastable-baselines
|
2022-11-07 12:46:22 +01:00 |
|
|
dodox
|
479d73ac4b
|
Hotfix for exploding gradients
|
2022-11-03 20:13:36 +01:00 |
|
|
dodox
|
82a174122a
|
Fixed bugs with givens-rotator for eigen
|
2022-10-24 10:08:31 +02:00 |
|
|
dodox
|
e31c25135e
|
Fixed Bug with Givens Rotation (Eigen)
|
2022-10-09 11:24:06 +02:00 |
|
|
dodox
|
f9a08add40
|
Action-Loss for SAC
|
2022-10-08 18:33:37 +02:00 |
|
|
dodox
|
3110275d7b
|
Fixed bug with EIGEN (par_type)
|
2022-10-07 11:38:34 +02:00 |
|
|
dodox
|
f37d3215a6
|
Bug Fix for Full Cov
|
2022-09-23 23:06:19 +02:00 |
|
|
dodox
|
00dbc9bdd8
|
Error when calculating action_loss
|
2022-09-12 22:28:57 +02:00 |
|
|
dodox
|
2c14edd3b0
|
Fix: Dependency
|
2022-09-03 23:41:06 +02:00 |
|
|
dodox
|
d3fa3cc997
|
Fixed Givens Dependency
|
2022-09-03 23:16:54 +02:00 |
|
|
dodox
|
6e1a7cecd5
|
Implemented correct clipping (from original SAC)
|
2022-09-03 13:46:17 +02:00 |
|
|
dodox
|
ee4a0eed56
|
Fixed SAC+SDE+SDC bugs
|
2022-09-03 13:08:31 +02:00 |
|
|
dodox
|
4532135812
|
Finalized factoring out projections
|
2022-09-03 11:59:16 +02:00 |
|
|
dodox
|
0aeea4e2e5
|
Fixed Bug: Wrong dimensions for action_loss
|
2022-09-03 11:44:01 +02:00 |
|
|
dodox
|
4bb772a251
|
Factor Projections out into metastable-projections
|
2022-09-03 11:37:41 +02:00 |
|
|
dodox
|
0a037deccc
|
Implemented cov parametrization via eigen-decomp
|
2022-09-03 11:16:41 +02:00 |
|
|
dodox
|
e4a8cfc349
|
Implemented action_loss
|
2022-09-03 11:16:29 +02:00 |
|
|
dodox
|
2f05474091
|
Fixed a bug with KL-proj
|
2022-08-28 20:48:02 +02:00 |
|
|
dodox
|
4080ad8135
|
Removed old TODOs
|
2022-08-28 12:07:19 +02:00 |
|
|
dodox
|
eb881559d6
|
Support clip_range None
|
2022-08-28 02:07:18 +02:00 |
|
|
dodox
|
1d3c2fe005
|
Allow completely disabling some PPO features (for TRPL)
|
2022-08-28 00:26:44 +02:00 |
|
|
dodox
|
afec4e709c
|
Fixed bug in RolloutBuffer when using parallel envs
|
2022-08-27 16:02:40 +02:00 |
|
|
dodox
|
02e4ed1510
|
Added support for parallel envs
|
2022-08-27 15:19:00 +02:00 |
|
|
dodox
|
5c39be5ead
|
Testing Observables
|
2022-08-22 15:05:42 +02:00 |
|
|
dodox
|
c6a58b15dd
|
Fixing SDE bug
|
2022-08-22 14:19:40 +02:00 |
|
|
dodox
|
197de7997c
|
Fixed bug with SDE
|
2022-08-22 13:36:17 +02:00 |
|
|
dodox
|
a9e3f295b2
|
Fixed numerical issues with Wasserstein
|
2022-08-17 23:25:24 +02:00 |
|
|
dodox
|
9fffe048af
|
Fixed Spherical_Chol not accepting batches
|
2022-08-17 22:55:42 +02:00 |
|
|
dodox
|
86e6bfb65b
|
Higher epsilon to deal with numerical instabilities
|
2022-08-17 19:31:54 +02:00 |
|
|
dodox
|
64a7d5ec59
|
Guarante minimum epsilon when ensuring non-zero (CholNet)
|
2022-08-16 20:02:33 +02:00 |
|
|
dodox
|
d35c3d8520
|
Fixed all the bugs in TRPL
|
2022-08-15 16:55:17 +02:00 |
|
|
dodox
|
28d0c609bc
|
Fixed SDE: sampling had dimension mismatches
|
2022-08-14 20:09:10 +02:00 |
|
|
dodox
|
e1c59cffd0
|
Removed debug-print
|
2022-08-14 18:45:17 +02:00 |
|
|
dodox
|
639aae7f42
|
Testing SDE...
|
2022-08-14 18:42:45 +02:00 |
|
|
dodox
|
bb1f9ecf2b
|
Fixed UniversalGaussianDistribution lost SDE when cloning
|
2022-08-14 18:42:19 +02:00 |
|
|
dodox
|
0ee65e789b
|
Fixing sde's bugs
|
2022-08-14 16:10:22 +02:00 |
|
|
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 |
|