Commit Graph
6257 Commits
Author SHA1 Message Date
Barak MichenerandGitHub 05c4e3fb2a [build] Build wheels with manylinux2014 (#11621)
* necessary changes

* Split bazel install

* manylinux2014

* change references to manylinux2014

* Fix lint

* port alex's docker build changes

* fix config issue

* remove extra manylinux2010 requirement script

* revert SHA overwrite

* wip

* incompatible_linklibs

* fix nits
2020-11-03 19:36:32 -08:00
Ian RodneyandGitHub 9527220a86 [serve] Fix Controller Crashes on Win (#11792) 2020-11-03 16:54:16 -08:00
architkulkarniandGitHub 2ef707e440 Update advanced.rst (#11793) 2020-11-03 16:16:36 -08:00
Sven MikaandGitHub 5b788ccb13 [RLlib] Trajectory view API (prep PR for switching on by default across all RLlib; plumbing only) (#11717) 2020-11-03 12:53:34 -08:00
Ian RodneyandGitHub c3074f559c [serve] Split out metadata for checkpointing (#11533) 2020-11-03 12:41:24 -08:00
Philipp MoritzandGitHub 39ce0eadbe Ray PDB support (#11739) 2020-11-03 09:49:23 -08:00
Stephanie WangandGitHub 952b71dc94 Fix windows build (#11786) 2020-11-03 12:38:45 -05:00
d352feadf0 [Dashboard] Memory Page Loading Wheel (#11651)
* Switch memory view loading message over to a loading wheel to make UX less confusing.

* lint

Co-authored-by: Max Fitton <max@semprehealth.com>
2020-11-03 09:37:30 -08:00
Ameer Haj AliandGitHub 08e0e8311a [autoscaler] Fixing AWS instance types autofill (#11758) 2020-11-03 09:34:14 -08:00
Kai FrickeandGitHub f7b19c41e3 [tune] logger refactor part 1: move classes and utilities to own files (#11746)
* [tune] logger refactor part 1: move classes and utilities to own files

* Fix circular dependency

* Remove uneeded pretty print copy

* Apply suggestions from code review
2020-11-03 07:48:09 -08:00
desktableandGitHub 5af745c90d [RLlib] Implement the SlateQ algorithm (#11450) 2020-11-03 09:52:04 +01:00
Lara CodecaandGitHub e735add268 [RLlib] Integration with SUMO Simulator (#11710) 2020-11-03 09:45:03 +01:00
0a6d24a727 [cli] Remove the deprecated old_style logging calls (#10776)
Co-authored-by: Richard Liaw <rliaw@berkeley.edu>
2020-11-02 23:40:18 -08:00
e7f7cb29c4 [docs] Show expected terminal output for manual cluster setup (#11752)
Co-authored-by: Richard Liaw <rliaw@berkeley.edu>
2020-11-02 20:59:14 -08:00
Ian RodneyandGitHub 6e89702508 [docker] Disable Readme push to avoid errors (#11770) 2020-11-02 19:12:51 -08:00
Max FittonandGitHub 3202ff74c2 [Dashboard] Don't show GPU columns if no GPU in cluster (#11704) 2020-11-02 18:07:27 -06:00
Stephanie WangandGitHub 0ba777af99 [Object spilling] Add policy to automatically spill objects on OutOfMemory (#11673) 2020-11-02 12:42:02 -08:00
Ameer Haj AliandGitHub 8d74a04a42 [autoscaler] Flag flip for resource_demand_scheduler should take into account queue (#11615) 2020-11-02 12:41:22 -08:00
cce91b51bd [docker] Fix docker regex (#11726)
Co-authored-by: Alex Wu <alex@anyscale.com>
2020-11-02 11:23:06 -08:00
Ian RodneyandGitHub 171e02c684 [serve] re-enable serve-controller-crash test (#11579) 2020-11-02 11:22:09 -08:00
4a7d0e059d [GCS]Optimize subscription perf (#11669)
* [GCS]Optimize subscription perf

* fix review comment

Co-authored-by: 灵洵 <fengbin.ffb@antfin.com>
2020-11-02 09:46:04 -08:00
dHannaschandGitHub 8346dedc3a Fix the linter failure. (#11755) 2020-11-02 18:02:15 +01:00
26176ec570 [RLlib] Fix epsilon_greedy on nested_action_spaces only in pytorch (#11453)
* [RLlib] Fix epsilon_greedy on nested_action_spaces only in pytorch

* epsilon_greedy on Continuous action

* formatt

* Fix error

* fix format

* fix bug

* increase speed

* Update rllib/utils/exploration/epsilon_greedy.py

* Update rllib/utils/exploration/epsilon_greedy.py

* Update rllib/utils/exploration/epsilon_greedy.py

Co-authored-by: Sven Mika <sven@anyscale.io>
2020-11-02 12:22:33 +01:00
Sven MikaandGitHub 54d85a6c2a [RLlib] Fix RNN learning for tf-eager/tf2.x. (#11720) 2020-11-02 11:18:41 +01:00
Sven MikaandGitHub bfc4f95e01 [RLlib] Fix test_bc.py test case. (#11722)
* Fix large json test file.

* Fix large json test file.

* WIP.
2020-10-31 00:16:09 -07:00
Eric LiangandGitHub 48dee789b3 Add random actor placement; fix cancellation callback; update test skips (#11684) 2020-10-30 18:36:35 -07:00
DK.PinoandGitHub b10871a1f5 [Core]Fix get workrer table bug (#11516)
* fix get_worker_table bug

* fix lint

* fix comment

* remove actor table

* fix comment

* fix get alive worker

* remove unused python import
2020-10-30 14:48:29 -07:00
SangBin ChoandGitHub 71c5089854 [Object Spilling] Initial Iteration of S3 adapter. (#11379)
* Finished the first iteration.

* Removed unnecessary code.

* Smartopen impl.

* Make sure tests passed.

* Addressed code review.

* Addressed code review.

* Fix issues.

* Fix issues.
2020-10-30 14:47:07 -07:00
Ameer Haj AliandGitHub 7aade469d0 [autoscaler] fix the autoscaling bug for continuously launching failed nodes (#11714) 2020-10-30 14:12:06 -07:00
Gekho457andGitHub 8816d34541 Kubernetes rsync verbosity fixed (#11716) 2020-10-30 14:03:42 -07:00
Alan GuoandGitHub 3c109b45aa Disable validation of cluster config on the cluster to allow for cluster configs with new properties. (#11693) 2020-10-30 14:02:00 -07:00
Eric LiangandGitHub f9f372c327 [autoscaler] Clean up monitoring loop code (#11677) 2020-10-30 13:48:43 -07:00
SangBin ChoandGitHub 6e2a1eac36 [Placement Group] Placement group automatic cleanup. (#11546)
* In progress. Done with all placement group manager code.

* It is working with job.

* Finished detached actor implementation.

* Fix minor issue.

* In progress.

* Addressed code review.

* Addressed code review.

* Addressed code reivew.

* Fix a build error.
2020-10-30 10:55:43 -07:00
5a83d8918a [release] Do not tag docker latest on release builds (#11694)
* fix

* Added comment

Co-authored-by: Alex Wu <alex@anyscale.com>
2020-10-29 23:13:25 -07:00
Max FittonandGitHub b4df42b027 [Dashboard] Make Infeasible Actor UX Less Scary (#11654)
* Update infeasible actor UI so that it only shows infeasible for an ActorClassGroup if at least one actor in the class is infeasible

* lint
2020-10-29 23:12:43 -07:00
Max FittonandGitHub d6628cdbfb [Dashboard] Fix null gpu utilization (#11650)
* update dashboard to work if GPU utilization field is missing from GPU payload

* lint

* lint
2020-10-29 23:11:50 -07:00
e022d12dc3 [New scheduler] Deflake test heartbeat (#11586)
* defleked

* lint

* .

* Update cluster_task_manager_test.cc

Co-authored-by: Alex Wu <alex@anyscale.com>
2020-10-29 23:10:19 -07:00
architkulkarniandGitHub 4175569d96 [Core] Add option to override environment variables for tasks and actors (#11619) 2020-10-29 14:22:44 -05:00
Simon MoandGitHub e82ff08b0c Fix asyncio plasma integration in cluster mode (#11665) 2020-10-29 11:53:10 -07:00
Lingxuan ZuoandGitHub 0b7a3d9e02 [Log] new spdlog tool for ray (#10967)
* spdlog support

* fatal abort for spdlog

* print all logs in stderr if no logger given

* fix log test

* install signal handler for spdlog by reusing glog lib

* fix lint

* Avoid duplicated dump

* log rotation and fmt comments

* fix
2020-10-29 11:37:13 -07:00
Ian RodneyandGitHub 87e971bff0 [docker] Include python k8s package in ray-deps (#11703) 2020-10-29 10:57:23 -07:00
Yutai ZhouandGitHub 6999db93cb Un-indent multiagent section (#11310)
* Un-indent multiagent section
MARL section used to be nested inside bandits, which we probably don't want. Maybe give it its own section instead?
2020-10-29 16:12:48 +01:00
0b07af374a allow tuple action space (#11429)
Co-authored-by: Jiajie Xiao <jj@Jiajies-MBP-2.attlocal.net>
2020-10-29 16:05:38 +01:00
Barak MichenerandGitHub 91fa7e0b4e [releng]: Quiet Docker Push (and explain why) (#11623) 2020-10-29 00:18:51 -07:00
Simon MoandGitHub 46afec5660 Mute asyncio warning for Serve (#11682) 2020-10-28 17:05:42 -07:00
huyz-gitandGitHub 64e3c9741a Update rllib-algorithms.rst (#11642) 2020-10-28 15:07:10 -07:00
mvindiola1andGitHub 9e68b77796 [RLLIB] Wait for remote_workers to finish closing environments before terminating (#11476) 2020-10-28 14:23:06 -07:00
Edward OakesandGitHub fcaf4d80e3 [serve] Make fractional resource usage more obvious in docs (#11580) 2020-10-28 13:54:36 -07:00
Kai FrickeandGitHub ba63ded311 [tune] better error when metric or mode unset in search algorithms (#11646) 2020-10-28 13:17:59 -07:00
Richard LiawandGitHub 58891551d3 [tune] make tests faster + fix flaky test (#10264) 2020-10-28 13:14:54 -07:00