wassname/kair_algorithms_draft
1
0
Fork 0
mirror of https://github.com/wassname/kair_algorithms_draft.git synced 2026-08-21 11:16:38 +08:00
Code Issues Packages Projects Releases Wiki Activity
32 Commits 11 Branches 0 Tags
master
Commit Graph
15 Commits
Author SHA1 Message Date
Donghyeon Kim 18d6824191 Refactoring demo collector (#56) 2019-08-17 14:29:58 +09:00
Kyunghwan Kim ede6fa3c7a Add HER on sac (#85) 
* Add HER on sac

* Add args in wandb log
 2019-07-20 17:36:01 +09:00
Whi Kwon 4b56ae781b Modify docker related commands (#54) 
* Add new ros package installation(orocos, hrl-kdl)

* Add source bashrc

* update train script

* Add custom package ros

* Change delegating env_name logic

* Add reacher env when training selection

* Change clone branch to master
 2019-05-21 13:26:52 +09:00
Whi Kwon c1f348775e Add openmanipulator simulation environment agent (#50) 2019-05-03 21:25:15 +09:00
Whi Kwon 16ae4375c9 Change demo file compatible to python2 (#40) 
* Change demo file to python2 compatible

* Add object to classes for compatibility with python2
 2019-04-07 13:06:39 +09:00
Kyunghwan Kim d2769dfa9d Convert code to python 2.7 (#35) 
* Convert code format to python2.7 (SAC)

* Convert code format python2.7 (TD3, all fD)

* Remove no use import and black setting

* Change SAC param

* Change env name Reacher-v2 to v1

* Remove old version reacher training script

* Convert code format python2.7

* Modify .travis.yml

* Add install command python3.6 & black on Makefile

* Fix seperator to tab on Makefile

* Modify Makefile

* Fix little error

* Change td3 gamma parameter
 2019-03-25 19:07:19 +09:00
Seungjae Ryan Lee 00c3dc4286 Remove DDPG (#28) 
* Change default algorithm from DDPG to SAC

* Remove DDPG examples

* Remove DDPG algorithms

DDPG, DDPG+PER, DDPGfD

* Fix parser bug in LunarLanderContinuous
 2019-03-21 23:51:39 +09:00
Kyunghwan Kim b4f8bd58b4 Add demo files (#23) 
* Add demo files

* Add demo path on run files and change formatting

* Delete discrete demo
 2019-03-14 14:02:24 +09:00
Seungjae Ryan Lee ca5c99bc41 Add DDPGfD, TD3fD and SACfD (#22) 
* Format repository

* Clone files from medipixel repo

* Fix DDPGfDAgent.update_model()

* Fix bug on _initialize()

* Add demo-path parameter and demo data

* Rename init_priority to _max_priority for PER

This makes PER and PERfD consistent.

* Make i_episode attribute of DDPGAgent

* Clone SAC code from medipixel repo

* Fix update_model() for SACfD

* Fix _initialize() for SACfD

* Add is_discrete attribute to AbstractAgent for SACfD

* Add i_episode attribute to SACAgent for SACfD

* Modularize DDPGAgent and SACAgent

* Modify hyperparameters for DDPGfD and SACfD

* Add NStepBuffer

* Add n-step to DDPGfD

* Add n-step to SACfD

* Add TD3fD without n-step

* Attempt to tune hyperparameters

* Remove discrete environment check in SAC

* Implement n-step on TD3fD

* Fix step function of TD3

No done check, and _add_transition_to_memory was not called.

* Fix actor loss calculation for TD3fD

* Attempt to tune hyperparameters

* Print both critic losses

* Fix typo bug

* Attempt to tune hyperparameters

* Fix bug in n-step demo retrieval

* Fix bug in n-step transition addition
 2019-03-14 11:06:54 +09:00
Whi Kwon d7b09d1bd5 Add LSTM network (#15) 
* Add common lstm module

* Add wave_test notebook

* Add lstm test example (sin -> cos)

* Add LSTM test case(sine+noise->cos)

* Move test_lstm file

* Add legend to distinguish label

* Fix flake8

* Add matplotlib to requirements
 2019-03-05 09:25:38 +09:00
Whi Kwon 5962ff5277 Add TD3 (#10) 
* Add td3

* Fix flake8

* Fix action clamping

* Increase episode max step, add detach to actor_loss

* Fix actor update freq bug

* Add per (#8)

* Add per and modify etc

* Replace pre-commit-config.yaml and add pre-commit hook in .git

* Modify .gitignore

* Modify .gitignore

* Modify buffer and code

* Modify replay buffer and per

* Modify .gitignore

* Add random initial action in ddpg (#13)

* Add random initial actions in ddpg

* Add reacher-v2 example of ddpg

* Add soft actor critic (#12)

* Add soft actor critic

* Delete unnecessary examples

* Add td3

* Fix flake8

* Fix action clamping

* Fix flake8

* Increase episode max step, add detach to actor_loss

* Fix actor update freq bug

* Fix code to reflect PR

* Resolve conflict
 2019-02-24 13:44:20 +09:00
Jinwoo Park (Curt) 4a19053d25 Add soft actor critic (#12) 
* Add soft actor critic

* Delete unnecessary examples
 2019-02-18 17:54:07 +09:00
Jinwoo Park (Curt) d2b670015c Add random initial action in ddpg (#13) 
* Add random initial actions in ddpg

* Add reacher-v2 example of ddpg
 2019-02-18 08:57:36 +09:00
Kyunghwan Kim ecb42d30d2 Add per (#8) 
* Add per and modify etc

* Replace pre-commit-config.yaml and add pre-commit hook in .git

* Modify .gitignore

* Modify .gitignore

* Modify buffer and code

* Modify replay buffer and per

* Modify .gitignore
 2019-02-16 17:30:12 +09:00
Whi Kwon 7f4756a1d4 Add overall setting and ddpg baseline (#1) 
* Add overall CI settings

* Add specific build dir to travis

* Add before install/script condition to travis

* Add ddpg baseline

* Add wandb, remove algorithms except ddpg

* Remove init file in script

* Separate config file for ddpg

* Remove unnecessary examples

* Remove unnecessary args opt

* Add pre-commit setting

* Change pre-commit settings

* Change travis-ci setting

* Fix travis-ci issue

* Modify argparse arguments, fix requirements

* Change arguments order
 2019-02-05 20:07:46 +09:00