gryf/coach
1
0
Fork 0
mirror of https://github.com/gryf/coach.git synced 2025-12-18 11:40:18 +01:00
Code Issues Projects Releases Wiki Activity
470 Commits 4 Branches 2 Tags
30c64d065628869e63d319e74c03ee053b79187d
Commit Graph

275 Commits

Author SHA1 Message Date
Gal Leibovich
d6795bd524 batchnorm fixes + disabling batchnorm in DDPG (#353) 
Co-authored-by: James Casbon <casbon+gh@gmail.com>
 2019-06-23 11:28:22 +03:00
anabwan
7b5d6a3f03 tests: stabling functional tests (#355) 
* tests: stabling functional tests

* functional removed
 2019-06-20 15:30:47 +03:00
shadiendrawis
8e812ef82f Coach as a library (#348) 
* CoachInterface + tutorial

* Some improvements and typo fixes

* merge tutorial 0 and 4

* typo fix + additional tutorial changes

* tutorial changes

* added reading signals and experiment path argument
 2019-06-19 18:05:03 +03:00
Gal Leibovich
7eb884c5b2 TD3 (#338) 2019-06-16 11:11:21 +03:00
Timo Kaufmann
8df3c46756 Do not hardcode path to bash (#332) 2019-06-10 20:10:28 +03:00
Gal Leibovich
a1bb8eef89 DDPG Critic Head Bug Fix (#344) 
* A bug fix for DDPG, where the update to the policy network was based on the sum of the critic's Q predictions on the batch instead of their mean
 2019-06-05 17:47:56 +03:00
anabwan
0aa5359d63 tests: added assert for cp param and changing test args order (#342) 2019-06-05 00:16:50 +03:00
Gal Leibovich
4c996e147e applying filters for a csv loaded dataset + some bug-fixes in data loading (#319) 2019-05-28 15:44:55 +03:00
anabwan
f5ba14575c tests: print logs on failure + fix -cp param (#327) 
* tests: pring logs on failure

* fix import

* added job to circleci

* fix functional

* removed debug job
 2019-05-28 13:45:43 +03:00
Gal Leibovich
251dc9ccc0 Preset dependent number of csv read attempts in golden testing (#334) 2019-05-28 12:19:57 +03:00
Gal Leibovich
9e9c4fd332 Create a dataset using an agent (#306) 
Generate a dataset using an agent (allowing to select between this and a random dataset)
 2019-05-28 09:34:49 +03:00
anabwan
342b7184bc Enabling Coach Documentation to be run even when environments are not installed (#326) 2019-05-27 10:46:07 +03:00
James Casbon
2b7d536da4 Add head regularization costs to tf.losses (#292) 2019-05-26 17:15:42 +03:00
anabwan
3b6e413532 tests: fix traces and changing workflow jobs (#316) 
* tests: fix traces export presets

* tests: increase time for traces

* tests

* remove approval

* fix approval

* fix ap

* change worflow jobs

* fix path

* fix repo path

* change run traces

* adding assert

* fix assert
 2019-05-26 15:27:36 +03:00
anabwan
b567091d2e removed timestep_limit due to gym version upgrade (#325) 
* removed timestep_limit due to gym version update

* removed _past_limit wrapper
 2019-05-26 13:58:16 +03:00
Gal Leibovich
30c2b2fc45 moving to skimage.transform.resize (#321) 2019-05-23 13:38:01 +03:00
Gal Leibovich
acceb03ac0 bug fixes for OPE (#311) 2019-05-21 16:39:11 +03:00
Gal Leibovich
deb0251367 bug fix following PR #191 (#313) 2019-05-12 13:42:45 -07:00
Gal Novik
aa9f3cefaf Printing input size as part of network summary (#310) 2019-05-12 15:40:02 +03:00
anabwan
ffb55b4142 tests: update traces (#302) 
* Traces folder removed from repo and moved to S3
* Traces jobs and update will use directly the S3 files
 2019-05-07 10:04:05 +03:00
anabwan
740359587d tests: fixed nightly (#301) 
* tests: fixed nightly

* tests: temp testing functional tests

* tests: temp testing functional tests

* tests: add seed to -cp

* test: last fix
 2019-05-05 08:28:57 +03:00
Gal Leibovich
582921ffe3 OPE: Weighted Importance Sampling (#299) 2019-05-02 19:25:42 +03:00
guyk1971
74db141d5e SAC algorithm (#282) 
* SAC algorithm

* SAC - updates to agent (learn_from_batch), sac_head and sac_q_head to fix problem in gradient calculation. Now SAC agents is able to train.
gym_environment - fixing an error in access to gym.spaces

* Soft Actor Critic - code cleanup

* code cleanup

* V-head initialization fix

* SAC benchmarks

* SAC Documentation

* typo fix

* documentation fixes

* documentation and version update

* README typo
 2019-05-01 18:37:49 +03:00
Ajay Deshpande
33dc29ee99 Uploading checkpoint if crd provided (#191) 
* Uploading checkpoint if crd provided
* Changing the calculation of total steps because of a recent change in core_types

Fixes #195
 2019-04-26 12:27:33 -07:00
anabwan
b3db9ce77d tests: fixed failed tests - stabling CI (#298) 
* tests: stabling CI

* tests: fix failed tests - stabling CI

* fix get csv files.
  - fixed seed test
* fix clres on conftest - now can modify paths during test run.
  - this fixed the mxnet checkpoint test

* tests: fix comments
 2019-04-23 15:12:11 +03:00
Gal Leibovich
9f625c197b fix for fetch rendering (#297) 
* fix for fetch rendering - removing code which was once required with older gym versions. images are now rendered correctly by default with the latest gym.

* fixing mujoco camera id failure
 2019-04-21 17:37:14 +03:00
Gal Leibovich
4741b0b916 BCQ variant on top of DDQN (#276) 
* kNN based model for predicting which actions to drop
* fix for seeds with batch rl
 2019-04-16 17:06:23 +03:00
Federico Andres Lois
bdb9b224a8 Include missing RegressionHead. (#263) 2019-04-16 15:24:06 +03:00
anabwan
20a8dea0dd tests: minor fix for functional tests (#289) 
* tests: minor fix for functional tests

* tests: fix value
 2019-04-15 12:28:23 +03:00
zach dwiel
88f9c926ab update comment describing why the output filters don't modify Agent.last_action_info 2019-04-09 12:14:27 -04:00
zach dwiel
fd2c210915 rename AgentInterface.emulate_observe_on_trainer or observe_transition and call from AgentInterface.observe 2019-04-09 12:14:27 -04:00
zach dwiel
f8741522e4 merge AgentInterface.emulate_act_on_trainer and AgentInterface.act 2019-04-09 12:14:27 -04:00
zach dwiel
f2fead57e5 change method interface: AgentInterface.emulate_act_on_trainer(transition: Transition) -> emulate_act_on_trainer(action: ActionType) 2019-04-09 12:14:27 -04:00
zach dwiel
b20e795ce0 create method LevelManager.acting_agent() 2019-04-09 12:14:27 -04:00
zach dwiel
54fdfe2da8 simplify rollout worker steps with new magic methods on StepMethod 2019-04-09 12:14:27 -04:00
zach dwiel
2cb078b4c2 add __truediv__, __rtruediv__ and __eq__ to StepMethod 2019-04-09 12:14:27 -04:00
zach dwiel
83da5cde2f remove unnecessary parentheses 2019-04-09 12:14:27 -04:00
zach dwiel
dddaefb210 fixed bug in rollout worker where total number of improved steps are not taken 2019-04-09 12:14:27 -04:00
zach dwiel
06de3b0f07 update LevelManager type signature 2019-04-09 12:14:27 -04:00
zach dwiel
f16cd3cb1e remove unused ActionInfo.action_intrinsic_reward 2019-04-09 12:14:27 -04:00
zach dwiel
7d79433c05 remove unused parameter scale_external_reward_by_intrinsic_reward_value 2019-04-09 12:14:27 -04:00
anabwan
881f78f45a tests: new checkpoint mxnet test + fix utils (#273) 
* tests: new mxnet test + fix utils

new test added:
- test_restore_checkpoint[tensorflow, mxnet]

fix failed tests in CI
improve utils

* tests: fix comments for mxnet checkpoint test and utils
 2019-04-07 07:36:44 +03:00
Zach Dwiel
2291cee2c6 allow serializing from/to arrays/str from GlobalVariableSaver (#285) 2019-04-04 11:09:19 -04:00
anabwan
cdb8d9e518 tests: fix multi environment variables in configci (#284) 
* tests: fix multi environment variables in configci

- fix multi environment vairables in configci
- removing bitflip from mujoco tests
- add bitflip to gym

* tests: disable mujoco_a3c_lstm + fix timeout and fix docker
 2019-04-04 16:11:41 +03:00
Scott Leishman
f173e69187 introduce dockerfiles. (#169) 
* introduce dockerfiles.

* ensure golden tests are run not just collected.

* Skip CI download of dockerfiles.

* add StarCraft environment and tests.

* add minimaps starcraft validation parameters.

* Add functional test running (from Ayoob)

* pin mujoco_py version to a 1.5 compatible release.

* fix config syntax issue.

* pin remaining mujoco_py install calls.

* Relax pin of gym version in gym Dockerfile.

* update makefile based on functional test filtering.
 2019-04-03 19:33:17 +03:00
shadiendrawis
0b808f0794 remove -ept flag (#283) 2019-04-03 16:32:24 +03:00
anabwan
869bd421a3 tests: added new checkpoint and functional tests (#265) 
* added new tests
- test_preset_n_and_ew
- test_preset_n_and_ew_and_onnx

* code utils improvements (all utils)
* improve checkpoint_test
* new functionality for functional_test markers and presets lists
* removed special environment container
* add xfail to certain tests
 2019-03-28 13:57:31 -07:00
Gal Leibovich
310d31c227 integration test changes to reach the train part (#254) 
* integration test changes to override heatup to 1000 steps +  run each preset for 30 sec (to make sure we reach the train part)

* fixes to failing presets uncovered with this change + changes in the golden testing to properly test BatchRL

* fix for rainbow dqn

* fix to gym_environment (due to a change in Gym 0.12.1) + fix for rainbow DQN + some bug-fix in utils.squeeze_list

* fix for NEC agent
 2019-03-27 21:14:19 +02:00
Gal Leibovich
6e08c55ad5 Enabling-more-agents-for-Batch-RL-and-cleanup (#258) 
allowing for the last training batch drawn to be smaller than batch_size + adding support for more agents in BatchRL by adding softmax with temperature to the corresponding heads + adding a CartPole_QR_DQN preset with a golden test + cleanups
 2019-03-21 16:10:29 +02:00
Gal Leibovich
abec59f367 fixes to rainbow dqn + a cartpole based golden test (#253) 2019-03-21 12:57:56 +02:00