Commits

Breadcrumbs

History for

ondb863aa

User selector

All users

Datepicker

All time

Commit History

Commits on Oct 26, 2021

• [RLlib] Unify the way we create local replay buffer for all agents (ray-project#19627)

  Show description for 99a0088
  

  gjoliver

  authored
  99a0088

  Copy full SHA for 99a0088

• [rllib] Add deterministic test to gpu (ray-project#19306)

  Show description for ad87ddf
  

  avnishn

  and

  sven1977

  authored
  ad87ddf

  Copy full SHA for ad87ddf

Commits on Oct 25, 2021

• [RLlib] Fix failing test cases: Soft-deprecate ModelV2.from_batch (in favor of ModelV2.__call__). (ray-project#19693)

  sven1977

  authored
  b213565

  Copy full SHA for b213565

• [RLlib] Some minor cleanups (buffer buffer_size -> capacity and others). (ray-project#19623)

  

  gjoliver

  authored
  89fbfc0

  Copy full SHA for 89fbfc0

• [RLlib] Added LearningRateSchedule and EntropyCoeffSchedule to TF and Torch versions of A3C and PPO (ray-project#19276)

  roireshef

  authored
  9b0352f

  Copy full SHA for 9b0352f

• [RLlib] clean up all the SampleBatch['is_training'] deprecation warnings (ray-project#19652)

  Show description for c3c4227
  

  gjoliver

  authored
  c3c4227

  Copy full SHA for c3c4227

Commits on Oct 22, 2021

• [Tune] Remove queue_trials. (ray-project#19472)

  xwjiang2010

  authored
  a632cb4

  Copy full SHA for a632cb4

Commits on Oct 20, 2021

• [rllib] Add entropy_coeff_schedule support for APPO. (ray-project#19544)

  Show description for 44a4e42
  

  gjoliver

  authored
  44a4e42

  Copy full SHA for 44a4e42

Commits on Oct 14, 2021

• [release tests] Switch from yaml.load to yaml.safe_load (ray-project#19365)

  carlogrisetti

  authored
  5cee8a1

  Copy full SHA for 5cee8a1

Commits on Oct 12, 2021

• [RLlib] Check training_enabled on PolicyServer (ray-project#19007)

  antoine-galataud

  authored
  edb338f

  Copy full SHA for edb338f

• [RLlib] Report timesteps_this_iter to Tune, so it can track/checkpoint/restore total timesteps trained. (ray-project#19264)

  Show description for 9226f9b
  

  gjoliver

  authored
  9226f9b

  Copy full SHA for 9226f9b

Commits on Oct 10, 2021

• [RLlib] Support >1 loss terms and optimizers for framework=tf2 (already supported for framework=[tf|torch]) (ray-project#19269)

  sven1977

  authored
  bd2d207

  Copy full SHA for bd2d207

Commits on Oct 8, 2021

• [RLlib] TF2/eager memory leak fixes. (ray-project#19198)

  sven1977

  authored
  d439fd7

  Copy full SHA for d439fd7

Commits on Oct 7, 2021

• [RLlib] Issue 18280: A3C/IMPALA multi-agent not working. (ray-project#19100)

  sven1977

  authored
  c3e3fc7

  Copy full SHA for c3e3fc7

• [RLlib] Issue 18104: Cannot set remote_worker_envs=True for non local-mode and MultiAgentEnv. (ray-project#19133)

  sven1977

  authored
  fd438d5

  Copy full SHA for fd438d5

Commits on Oct 6, 2021

• [RLlib] Issue 18418: SAC w/ dict space not working. (ray-project#19101)

  sven1977

  authored
  1f0646f

  Copy full SHA for 1f0646f

Commits on Oct 4, 2021

• [RLlib] Issue 18812: Torch multi-GPU stats not protected against race conditions. (ray-project#18937)

  sven1977

  authored
  b4300dd

  Copy full SHA for b4300dd

• [RLlib] Fix flakey test_a3c, test_maml, test_apex_dqn. (ray-project#19035)

  sven1977

  authored
  73f5c40

  Copy full SHA for 73f5c40

• [Lint] Add flake8-bugbear (ray-project#19053)

  Show description for 7588bfd
  jjyao

  authored
  7588bfd

  Copy full SHA for 7588bfd

Commits on Sep 30, 2021

• [RLlib] Fix broken test_r2d2.py. (ray-project#19017)

  sven1977

  authored
  16ad46a

  Copy full SHA for 16ad46a

• [RLlib] Discussion 3644: Fix bug for complex obs spaces containing Box([2D shape]) and discrete component. (ray-project#18917)

  sven1977

  authored
  ac3371a

  Copy full SHA for ac3371a

• [RLlib] Unify all RLlib Trainer.train() -> results[info][learner][policy ID][learner_stats] and add structure tests. (ray-project#18879)

  sven1977

  authored
  ed85f59

  Copy full SHA for ed85f59

• [RLlib] Custom view requirements (e.g. for prev-n-obs) work with compute_single_action and compute_actions_from_input_dict. (ray-project#18921)

  sven1977

  authored
  828f5d2

  Copy full SHA for 828f5d2

• [RLlib] Raise error for kl penalty ddpo (ray-project#18959)

  Show description for 6dc1a6b
  

  avnishn

  and

  sven1977

  authored
  6dc1a6b

  Copy full SHA for 6dc1a6b

• [RLlib] Issue 18668: Unity3D env client/server example not working (fix + add to test cases). (ray-project#18942)

  sven1977

  authored
  05a55a9

  Copy full SHA for 05a55a9

Commits on Sep 29, 2021

• [RLlib] Allow n-step > 1 and prio. replay for R2D2 and RNNSAC. (ray-project#18939)

  sven1977

  authored
  9c9b482

  Copy full SHA for 9c9b482

• [RLlib] Add support for IMPALA to handle more than one loss/optimizer (analogous to recent enhancement for APPO). (ray-project#18971)

  sven1977

  authored
  b999438

  Copy full SHA for b999438

Commits on Sep 24, 2021

• [RLlib] Add unit tests for updating episode data in base_env (ray-project#17137)

  mvindiola1

  authored
  62f5da0

  Copy full SHA for 62f5da0

• [RLlib] Fix train.py input config patching (ray-project#18747)

  juliusfrost

  authored
  8b8b447

  Copy full SHA for 8b8b447

• [RLlib] Attention Nets + MultiDiscrete spaces: Fix range() takes no keyword args error! (ray-project#17502)

  o0olele

  authored
  ff6730f

  Copy full SHA for ff6730f

Commits on Sep 23, 2021

• [RLlib] No Preprocessors (part 2). (ray-project#18468)

  sven1977

  authored
  61a1274

  Copy full SHA for 61a1274

• [RLlib] Faster remote worker space inference (don't infer if not required). (ray-project#18805)

  sven1977

  authored
  a2a077b

  Copy full SHA for a2a077b

• [RLlib] Reinstate trajectory view API tests. (ray-project#18809)

  sven1977

  authored
  a96dbd8

  Copy full SHA for a96dbd8

Commits on Sep 22, 2021

• [RLlib] Increase size of (very flakey) action_masking example script test. (ray-project#18816)

  sven1977

  authored
  93208bb

  Copy full SHA for 93208bb

Commits on Sep 21, 2021

• [RLlib] POC: Separate losses for APPO/IMPALA. Enable TFPolicy to handle multiple optimizers/losses (like TorchPolicy). (ray-project#18669)

  sven1977

  authored
  698b4ee

  Copy full SHA for 698b4ee

Pagination

Previous

Next