hongshaorou/Tianshou
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Commit Graph

  • 023b33c917 Make mypy happy Dominik Jain 2023-10-13 12:25:28 +02:00
  • 76e870207d Improve persistence handling Dominik Jain 2023-10-12 17:40:16 +02:00
  • ba803296cc Add FileLoggerContext Dominik Jain 2023-10-12 17:39:11 +02:00
  • 3691ed2abc Support obs_rms persistence for MuJoCo by adding a general mechanism for attaching persistence to Environments instances Dominik Jain 2023-10-12 15:01:49 +02:00
  • f6d49774a2 Reify policy persistence, introducing Wold representation Dominik Jain 2023-10-11 19:31:26 +02:00
  • ee3813b09c Ignore temp scripts and temp folder Dominik Jain 2023-10-12 13:10:54 +02:00
  • 686fd555b0 Extend tests, fixing some default behaviour Dominik Jain 2023-10-11 16:07:34 +02:00
  • a8a367c42d Support IQN in high-level API Dominik Jain 2023-10-11 15:31:38 +02:00
  • 213e08a846 Add method get_output_dim to BaseActor Dominik Jain 2023-10-11 15:29:47 +02:00
  • c7d0b6b4b2 Simplify agent factories by making better use of base classes Dominik Jain 2023-10-10 20:07:30 +02:00
  • 799beb79b4 Support discrete SAC in high-level API Dominik Jain 2023-10-10 19:11:49 +02:00
  • 305b30a6c1 Simplify parameter transformers by applying ParamTransformerChangeValue Dominik Jain 2023-10-10 16:12:29 +02:00
  • 17ef4dd5eb Support REDQ in high-level API Dominik Jain 2023-10-10 15:49:05 +02:00
  • 7af836bd6a Support TRPO in high-level API and add example mujoco_trpo_hl Dominik Jain 2023-10-10 14:14:00 +02:00
  • 383a4a6083 Support NPG in high-level API and add example mujoco_npg_hl Dominik Jain 2023-10-10 13:47:30 +02:00
  • 73a6d15eee Log Environments Dominik Jain 2023-10-10 13:26:07 +02:00
  • a8ea6808c3 Fix ruff type comparison complaint Dominik Jain 2023-10-10 13:17:09 +02:00
  • 1bb52a6a5c Simplify critic/agent with optimizer generation Dominik Jain 2023-10-10 13:12:25 +02:00
  • 6bb3abb2f0 Support PG/Reinforce in high-level API Dominik Jain 2023-10-10 12:55:25 +02:00
  • 4e93c12afa Remove obsolete configuration files Dominik Jain 2023-10-09 18:01:10 +02:00
  • 22dfc4ed2e Fix type annotations of dist_fn Dominik Jain 2023-10-09 17:48:43 +02:00
  • a161a9cf58 Improve type annotations, fix type issues and add checks Dominik Jain 2023-10-09 17:22:52 +02:00
  • e6716326bd Make mypy ignore copied util modules string & logging Dominik Jain 2023-10-09 17:35:02 +02:00
  • 7ed6c1d71c Remove obsolete module highlevel.utils Dominik Jain 2023-10-09 14:21:36 +02:00
  • 1243894eb8 Add DistributionFunctionFactory subclasses for discrete/continuous default Dominik Jain 2023-10-06 14:32:21 +02:00
  • a8dc75fbab ExperimentBuilder: Allow experiment_config and sampling_config to be None Dominik Jain 2023-10-06 13:57:00 +02:00
  • 837ff13c04 Reorder ExperimentBuilder args (EnvFactory first) Dominik Jain 2023-10-06 13:53:45 +02:00
  • d269063e6a Remove 'RL' prefix from class names Dominik Jain 2023-10-06 13:50:23 +02:00
  • 50ac385321 Add some basic tests for high-level experiment builder API Dominik Jain 2023-10-05 19:22:04 +02:00
  • b54fcd12cb Change high-level DQN interface to expect an actor instead of a critic, because that is what is functionally required Dominik Jain 2023-10-05 19:21:08 +02:00
  • 1cba589bd4 Add DQN support in high-level API Dominik Jain 2023-10-05 15:39:32 +02:00
  • 358978c65d Add ToStringMixin to further high-level parameter classes Dominik Jain 2023-10-05 13:15:24 +02:00
  • 8f67c2e9d9 Disable numba DEBUG logs Dominik Jain 2023-10-03 21:16:53 +02:00
  • 9f0a410bb1 Log full experiment configuration, adding string representations to relevant classes Dominik Jain 2023-10-03 21:14:22 +02:00
  • 58bd20f882 Add logging module Dominik Jain 2023-10-03 21:03:39 +02:00
  • ce26e25923 Handle ruff complaints in string module Dominik Jain 2023-10-03 20:59:06 +02:00
  • de70147752 Add string module from sensAI Dominik Jain 2023-10-03 20:40:49 +02:00
  • 2671580c6c Add DDPG high-level API and MuJoCo example Dominik Jain 2023-10-03 20:26:39 +02:00
  • 6b6d9ea609 Add support for discrete PPO Dominik Jain 2023-09-28 20:07:52 +02:00
  • e0e7349b0a Add base class BaseActor with method get_preprocess_net for high-level API Dominik Jain 2023-09-28 20:08:55 +02:00
  • cd79cf8661 Add A2C high-level API Dominik Jain 2023-09-28 14:28:03 +02:00
  • acd89fa3b0 Remove parameter transformers from config object state, composing the list dynamically instead Dominik Jain 2023-09-27 18:20:49 +02:00
  • 78b6dd1f49 Adapt class naming scheme Dominik Jain 2023-09-27 17:20:35 +02:00
  • 5bcf514c55 Add alternative functional interface for environment creation where a persistable configuration object is passed as an argument, as this can help to ensure persistability (making the requirement explicit) Michael Panchenko 2023-09-27 14:10:45 +02:00
  • d4e604b46e Move parameter transformation directly into parameter objects, achieving greater separation of concerns and improved maintainability Dominik Jain 2023-09-26 17:43:16 +02:00
  • 38cf982034 Disable Ruff rule D205 (blank-line-after-summary) because it disallows, in particular, class docstrings that consist only of a summary line Dominik Jain 2023-09-26 17:40:21 +02:00
  • e993425aa1 Add high-level API support for TD3 Dominik Jain 2023-09-26 15:35:18 +02:00
  • 6a739384ee WandbLogger: Use less restrictive type annotation for config Dominik Jain 2023-09-26 10:48:13 +02:00
  • 367778d37f Improve high-level policy parametrisation Dominik Jain 2023-09-25 17:56:37 +02:00
  • 37dc07e487 Add high-level experiment builder interface Dominik Jain 2023-09-21 12:36:27 +02:00
  • 4a51e69265
         Bump urllib3 from 2.0.6 to 2.0.7 (#972) dependabot[bot] 2023-10-17 21:13:19 -04:00
  • bf7841078d
         Fixed the mapolicy train issue (#968) Fahmid Morshed Fahid 2023-10-16 20:52:07 -04:00
  • 66b7fc542b
         Minor dep update (#961) Michael Panchenko 2023-10-09 22:10:09 +02:00
  • 4d53d345d6 Ignore Ruff rule RET505, because it sacrifices visual discernability of control flow paths for brevity (regarding return statements) Dominik Jain 2023-09-25 11:18:17 +02:00
  • 3fd60f9e70 Unify PPO configuration objects, use experiment-specific configuration in mujoco_ppo_hl Dominik Jain 2023-09-20 15:45:09 +02:00
  • 8ec42009cb Move RLSamplingConfig to separate module config, fixing cyclic import Dominik Jain 2023-09-20 15:28:33 +02:00
  • d26b8cb40c Use experiment-specific config in mujoco_sac_hl, adding auto-alpha Dominik Jain 2023-09-20 15:13:05 +02:00
  • adc324038a Remove LoggerConfig Dominik Jain 2023-09-20 15:10:19 +02:00
  • 997b520580 Refactoring, dropping package config Dominik Jain 2023-09-20 13:15:06 +02:00
  • 316eb3c579 Add SAC high-level interface Dominik Jain 2023-09-20 09:29:34 +02:00
  • 2a1cc6bb55 Enable ruff setting ignore-init-module-imports Dominik Jain 2023-09-20 11:27:00 +02:00
  • 25c6bbd38c Ignore D106: Missing docstring in public nested class Dominik Jain 2023-09-20 11:24:51 +02:00
  • 16ed5fd2a5 Initial high-level interfaces, demonstrated in mujoco_ppo_hl Dominik Jain 2023-09-19 18:53:11 +02:00
  • a54aade730 Addition of dataclasses based config for scripts, major refactoring Michael Panchenko 2023-07-26 20:24:33 +02:00
  • 42fc181d74 Add dev dependencies jsonargparse and docstring_parser Dominik Jain 2023-09-20 09:51:06 +02:00
  • b900fdf6f2
         Remove kwargs in policy init (#950) Michael Panchenko 2023-10-08 17:57:03 +02:00
  • bc7ec9c149
         Bump pillow from 10.0.0 to 10.0.1 (#958) dependabot[bot] 2023-10-03 20:31:57 -07:00
  • b24f270a74
         Bump urllib3 from 1.26.16 to 1.26.17 (#957) dependabot[bot] 2023-10-03 23:58:26 +00:00
  • d11a5a3d99
         Bump gitpython from 3.1.33 to 3.1.35 (#953) dependabot[bot] 2023-10-03 07:52:57 +00:00
  • c30b4abb8f
         Add calibration to CQL as in CalQL paper arXiv:2303.05479 (#915) Anas BELFADIL 2023-10-03 07:54:34 +02:00
  • 6449a43261
         Fix documentation build (#951) Jiayi Weng 2023-09-26 08:24:08 -07:00
  • c8e7d02cba
         Minor: use Self type where appropriate (#942) Michael Panchenko 2023-09-20 00:40:32 +02:00
  • 2cc34fb72b
         Poetry install, remove gym, bump python (#925) Michael Panchenko 2023-09-05 23:34:23 +02:00
  • 600f4bbd55
         Python 3.9, black + ruff formatting (#921) Michael Panchenko 2023-08-25 23:40:56 +02:00
  • 07702fc007
         Improved typing and reduced duplication (#912) Michael Panchenko 2023-08-22 18:54:46 +02:00
  • 80a698be52
         Custom keys support in ReplayBuffer (#903) Anas BELFADIL 2023-08-11 01:06:10 +02:00
  • 61182450b6
         add py.typed, drop 3.6/3.7, support 3.11 (#910) Jiayi Weng 2023-08-10 14:13:46 -07:00
  • 864ee3df2f
         Make monitor_gym configurable in WandbLogger. (#896) Błażej Osiński 2023-08-10 00:13:25 +02:00
  • cd218dc12d
         Add assert description. (#894) Błażej Osiński 2023-08-10 00:12:42 +02:00
  • cb8551f315
         Fix master branch test issues (#908) Anas BELFADIL 2023-08-09 19:27:18 +02:00
  • f8808d236f
         fix a problem of the atari dqn example (#861) Zhenjie Zhao 2023-04-30 23:44:27 +08:00
  • 7ce62a6ad4
         actor critic share head bug for example code without sharing head - unify code style (#860) Gen 2023-04-29 06:43:22 +02:00
  • 1423eeb3b2
         Add warnings for duplicate usage of action-bounded actor and action scaling method (#850) ChenDRAG 2023-04-24 07:03:31 +08:00
  • e7c2c3711e
         Update gail.py (#849) wckwan 2023-04-13 14:25:57 +00:00
  • 4ac407c78f
         Remove test_fn and train_fn as they are not used in PPO PistonBall example for PettingZoo (#840) Quoding 2023-03-31 13:43:21 -04:00
  • 7f8fa241dd
         making pettingzoo a core dep instead of optional req (#837) Jiayi Weng 2023-03-25 22:01:09 -07:00
  • d5d521b329
         fix conda installation command (#830) Jiayi Weng 2023-03-19 17:40:47 -07:00
  • efdf72cb31 fix sphinx itemlist render error Jiayi Weng 2023-03-12 22:27:39 -07:00
  • f0afdeaf6a
         update version to 0.5.0 (#826) v0.5.0 Jiayi Weng 2023-03-12 22:07:16 -07:00
  • 73600edc58
         fix a bug in batch._is_batch_set (#825) Oren Zeev-Ben-Mordehai 2023-03-13 01:58:09 +01:00
  • bc222e87a6
         Fix #811 (#817) sunkafei 2023-03-04 08:57:04 +08:00
  • c8be85b240 fix readthedocs build error Jiayi Weng 2023-02-03 14:55:53 -08:00
  • e8acf0dd46
         Fix readthedocs build failure (#803) Jiayi Weng 2023-02-03 14:40:05 -08:00
  • 6c6c872523
         Gymnasium Integration (#789) Markus Krimmel 2023-02-03 20:57:27 +01:00
  • 6019406cff
         Add "act" to preprocess_fn call in collector. (#801) Jose Antonio Martin H 2023-02-03 20:19:38 +01:00
  • 774d3d8e83
         Implement args/kwargs for init of norm_layers and activation (#788) janofsssun 2022-12-26 19:58:03 -08:00
  • 1037627a5b
         fix info not pass issue in PGPolicy (#787) v0.4.11 Jiayi Weng 2022-12-24 13:06:54 -08:00
  • 4c3791a459
         Updated atari wrappers, fixed pre-commit (#781) Markus Krimmel 2022-12-04 22:00:53 +01:00
  • 662af52820
         Fix Atari PPO example (#780) Yi Su 2022-12-04 12:23:18 -08:00
  • 929508ba77
         Update experiment details of MuJoCo benchmark (#779) ChenDRAG 2022-11-27 02:18:22 +08:00