saymrwulf/stable-baselines3
1
0
Fork 0
mirror of https://github.com/saymrwulf/stable-baselines3.git synced 2026-09-05 20:30:42 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions
309 commits 18 branches 32 tags 5.2 MiB
Commit graph

266 commits

Author SHA1 Message Date
Antonin Raffin
546fbca20c Copy by default 2020-02-03 15:42:14 +01:00
Antonin Raffin
f7af08bea4 Refactor buffers to use numpy 2020-02-03 15:40:34 +01:00
Antonin Raffin
e3c5b1621e Sync Monitor with Stable Baselines 2020-02-03 15:07:35 +01:00
Antonin Raffin
f0dba88275 Fix statement that had no effects 2020-02-03 10:51:29 +01:00
Antonin Raffin
6710f1576c Fix eval log path 2020-01-31 13:48:25 +01:00
Antonin Raffin
ec657cc34e Fix tests and change log_path behavior for EvalCallback 2020-01-31 13:42:04 +01:00
Antonin Raffin
5d4e73544c Fix reset_num_timesteps 2020-01-31 13:16:28 +01:00
Antonin Raffin
6d59bfd4a0 Merge branch 'master' into feat/callbacks 2020-01-31 13:09:55 +01:00
Dormann, Noah
1f0dd60b97 Fix saving on GPU - Loading on CPU (#45) 
* removed policy from save, changed th.loads to map to device

* found hack: catch pickle exception and trying th.load with mapping instead, otherwise raise exception with more information -> loading cuda on cpu raises exception -> leads to th.load with map being called

* deleted todo

* updated changelog

* start of saving refactor

* first working c

* all tests pass, save refactored

* - backwards compatibilty not always
- make pytest all passing
- make typing all passing

* Fixes and simplify the save method

* Remove unused param

* Fix backward compat

* Fix docstring
 2020-01-31 13:06:55 +01:00
Antonin Raffin
6ae8421615 Update docstring 2020-01-28 10:28:44 +01:00
Antonin Raffin
5c94a225ef Minor edit to the doc 2020-01-28 10:24:02 +01:00
Antonin Raffin
a628354721 Refactor evaluation 2020-01-27 15:53:27 +01:00
Antonin Raffin
b66003cfb3 Add callback support 2020-01-27 14:32:31 +01:00
Antonin Raffin
fb57a6b80c Update docstring 2020-01-23 11:20:12 +01:00
Antonin Raffin
7265d9e352 Fix multiline f-string 2020-01-23 10:56:53 +01:00
Antonin Raffin
ff0eddfb17 Partially type base class 2020-01-22 17:51:27 +01:00
Antonin Raffin
44fce7c02a Fix typing errors and typos 2020-01-22 17:17:12 +01:00
Antonin Raffin
88f07bafb6 Convert format to f-strings 2020-01-22 16:39:25 +01:00
Antonin Raffin
37f9f13684 Revert all changes for python 2 
+ Add makefile and pytype
 2020-01-22 16:18:27 +01:00
Antonin Raffin
9e250b6818 Build doc 2020-01-20 16:19:35 +01:00
Antonin Raffin
b8df12afe2 Release v0.1.0 2020-01-20 13:01:14 +01:00
Antonin Raffin
0bed698ec5 Raise error for abstract methods 2020-01-20 12:57:40 +01:00
Antonin Raffin
e5c6601726 Update VecNormalize (pickling) and improve tests 2020-01-20 11:58:16 +01:00
Antonin Raffin
89db65b1fb Improve logger testing + add readers 2020-01-20 11:58:00 +01:00
Antonin Raffin
c542009641 Clean up code + bug fixes 2020-01-20 11:17:55 +01:00
Antonin Raffin
ea20721632 Add TODO 2020-01-15 15:58:45 +01:00
Antonin Raffin
03e853997a Add squash_output and expln as policy param for ppo and a2c 2020-01-15 13:21:20 +01:00
Antonin Raffin
60d5f4463d Add use_expln option for td3 2020-01-08 17:04:28 +01:00
Antonin Raffin
d3a718b94e Add extra dependency 2020-01-08 11:26:57 +01:00
Antonin Raffin
299ca007b5 Add comment about warmup phase 2020-01-07 17:36:26 +01:00
Antonin Raffin
8831eff163 Unify evaluation 2020-01-07 14:00:03 +01:00
Antonin RAFFIN
aa7b91333e Add seeding for subproc vecenv 2019-12-30 12:01:37 +01:00
Antonin RAFFIN
4a79f7e5a7 Print std reward for evaluation 2019-12-24 13:12:04 +01:00
Antonin RAFFIN
57c890f3e9 LeakyClip not working yet 2019-12-22 14:38:30 +01:00
Antonin RAFFIN
3a7508ac16 Fix double clip 2019-12-22 13:56:30 +01:00
Antonin Raffin
f6c475a44b Add use_expln as a policy argument 2019-12-20 18:10:24 +01:00
Antonin Raffin
7f34108ed6 Fix exp_ln computation 2019-12-20 18:02:01 +01:00
Antonin Raffin
9b3b34c9c4 Sample batch_size noise matrices for SAC 2019-12-20 11:28:44 +01:00
Antonin Raffin
161c608f9c Re-sample noise matrix for PPO 2019-12-20 11:28:20 +01:00
Antonin Raffin
e894f1f11b Add leakyclip 2019-12-19 18:20:02 +01:00
Antonin Raffin
69428346bd Bump version 2019-12-19 15:32:03 +01:00
Antonin Raffin
84ebc3d7da Relax the HardTanh limit 2019-12-19 15:28:51 +01:00
Antonin Raffin
bff0ca0ea8 Use HardTanh to relax the constrain 2019-12-19 11:59:00 +01:00
Antonin Raffin
c05c990285 Remove norm clipping 2019-12-18 16:56:51 +01:00
Antonin Raffin
e49d97bf98 Fix infs in SAC by bounding the mean 2019-12-18 13:45:33 +01:00
Antonin Raffin
57708a628c Add value function for SDE + TD3 2019-12-17 15:01:08 +01:00
Antonin Raffin
1d6f9bf100 Add sample freq for SDE 2019-12-17 11:47:21 +01:00
Antonin Raffin
919dfee452 Try to clip grad norm 2019-12-17 11:14:44 +01:00
Antonin Raffin
d63cef7693 Add gradient clipping for SAC 2019-12-06 18:32:57 +01:00
Antonin Raffin
233f346d53 Update todos 2019-12-06 17:46:56 +01:00
Antonin Raffin
6c423add8d Bump version 2019-12-05 16:44:27 +01:00
Antonin Raffin
0117cc37f4 Merge branch 'master' into feat/sde-features 2019-12-05 16:33:41 +01:00
Noah Dormann
aa67147796 clarified bytesIO use for load 2019-12-05 15:45:05 +01:00
Raffin, Antonin
bac9d4efed Update torchy_baselines/common/base_class.py 2019-12-05 14:53:14 +01:00
Raffin, Antonin
695cdc63a4 Update torchy_baselines/common/base_class.py 2019-12-05 14:52:59 +01:00
Raffin, Antonin
424a554567 Update docstring 2019-12-05 14:50:11 +01:00
Raffin, Antonin
464dd773e6 Update comment 2019-12-05 14:46:02 +01:00
Raffin, Antonin
03ecb17ef6 Update error message 2019-12-05 14:41:39 +01:00
Noah Dormann
88d4f44d55 added set_env test and set_env wrapping 2019-12-05 13:59:07 +01:00
Noah Dormann
cf1d7118a5 replaced file with file_path 2019-12-05 13:44:02 +01:00
Noah Dormann
8062ed6036 fixed load, to check if environment ist correctly 2019-12-05 13:36:19 +01:00
Noah Dormann
4b1bab7f85 implemented set_env method 2019-12-05 09:11:30 +01:00
Noah Dormann
8460bfe397 added some comments to _load_from_file 2019-12-05 08:56:04 +01:00
Noah Dormann
6560ae9952 using other_file instead of other_files 2019-12-05 08:52:26 +01:00
Noah Dormann
7c8d375bcb added get_parameter_list function 2019-12-05 08:50:11 +01:00
Noah Dormann
ff7c4d24f4 deleted types in train_actor td3 2019-12-05 08:46:26 +01:00
Noah Dormann
4c0f6cbe53 update get_opt_parameters to remove duplicate code 2019-12-05 08:43:12 +01:00
Noah Dormann
c3b0398d56 Changed load so it still works when env not saved 
improved save function
 2019-12-05 08:40:28 +01:00
Dormann, Noah
bea2799691 comment refactoring 
Co-Authored-By: Raffin, Antonin <Antonin.Raffin@dlr.de>
 2019-12-05 08:10:39 +01:00
Dormann, Noah
a756f40223 comment refactoring 
Co-Authored-By: Raffin, Antonin <Antonin.Raffin@dlr.de>
 2019-12-05 08:10:25 +01:00
Dormann, Noah
fdb544e775 comment refactoring 
Co-Authored-By: Raffin, Antonin <Antonin.Raffin@dlr.de>
 2019-12-05 08:10:14 +01:00
Dormann, Noah
aa66d2f82e comment refactoring 
Co-Authored-By: Raffin, Antonin <Antonin.Raffin@dlr.de>
 2019-12-05 08:09:51 +01:00
Dormann, Noah
85d3743224 added standart exclude parameters 
Co-Authored-By: Raffin, Antonin <Antonin.Raffin@dlr.de>
 2019-12-05 08:09:26 +01:00
Dormann, Noah
03bf513e5e comment refactoring 
Co-Authored-By: Raffin, Antonin <Antonin.Raffin@dlr.de>
 2019-12-05 08:08:20 +01:00
Antonin Raffin
1ac1a7cad5 Reformat 2019-12-02 14:27:38 +01:00
Antonin Raffin
3cdd5f20af Bug fix + add test for sde net arch 2019-12-02 14:14:48 +01:00
Antonin Raffin
8e9802784c Revert previous changes in SAC + SDE 2019-12-02 14:06:30 +01:00
Antonin Raffin
4e39a0627c Refactor: enable sde net arch for TD3 and SAC 2019-12-02 14:06:17 +01:00
Antonin Raffin
a2a8bbdf11 Sample n matrices for A2C/PPO when using SDE 2019-12-02 11:48:34 +01:00
Antonin Raffin
7a6a500398 Allow to use states directly as features for sde 2019-12-02 11:48:16 +01:00
Antonin RAFFIN
879191b26a Bug fix in SAC with constant ent coeff + try batch sde matrices 2019-12-01 13:11:13 +01:00
Noah Dormann
6928879f5a Refactored doc-strings 2019-11-28 16:30:13 +01:00
Noah Dormann
7ce610fade Deleted exact match parameter of load_parameters 2019-11-28 16:11:35 +01:00
Noah Dormann
ee6f938ddc Added option to explicitly specify excluded parameters 2019-11-28 15:42:53 +01:00
Noah Dormann
e95858784a Formatted all files 2019-11-28 15:38:04 +01:00
Noah Dormann
751ccf85e7 _setup_model() is now called when model is loaded 2019-11-28 15:33:52 +01:00
Noah Dormann
9ff59eaf3d Added attribute self.policy_class to prevent errors when using self.policy as class 2019-11-28 15:25:01 +01:00
Noah Dormann
e26564e0ec Added function for setting up any attributes that weren't saved and thus not loaded 2019-11-28 13:35:16 +01:00
Noah Dormann
c75582dfbe resolving conflicts 
# Conflicts:
#	torchy_baselines/a2c/a2c.py
#	torchy_baselines/ppo/ppo.py

Added optimizer params test
 2019-11-28 12:12:06 +01:00
Noah Dormann
812cab84ac Changed PPO deterministic 2019-11-28 11:20:40 +01:00
Noah Dormann
b75ffe166a Cleared base_class.load description 2019-11-28 10:36:21 +01:00
Antonin Raffin
fbe29a7298 Track down autograd error 
"Trying to backward through the graph a second time" -> added a comment
 2019-11-27 17:29:47 +01:00
Antonin Raffin
fe67a98711 Log more values 2019-11-26 17:44:06 +01:00
Antonin Raffin
5483e02d1a Add SDE support for SAC 2019-11-26 15:26:12 +01:00
Antonin Raffin
0885dbe74b Bug fix in choosing the distribution 2019-11-25 15:02:10 +01:00
Antonin Raffin
5d6649d92b Enable separate feature extraction for SDE 2019-11-25 14:54:13 +01:00
Antonin Raffin
d0003ee4ec Enable kwargs for proba dist 2019-11-25 14:00:21 +01:00
Antonin Raffin
5bbb14188d Unify A2C and TD3 SDE implementation 2019-11-25 13:19:33 +01:00
Antonin Raffin
c56865e10d Cleanup CEM, rename variables + add comments 2019-11-22 19:02:00 +01:00
Antonin Raffin
c47be0086e Add docstrings 2019-11-22 17:24:47 +01:00