Commit Graph
322 Commits
Author SHA1 Message Date
Eric LiangandPhilipp Moritz 3896b726dd Dynamically adjust redis memory usage (#4152)
* f

* Update services.py
2019-02-25 16:21:37 -08:00
Eric LiangandGitHub d9da183c7d [rllib] Custom supervised loss API (#4083) 2019-02-24 15:36:13 -08:00
Robert NishiharaandPhilipp Moritz 7b04ed059e Move TensorFlowVariables to ray.experimental.tf_utils. (#4145) 2019-02-24 14:26:46 -08:00
Eric LiangandGitHub 05d96ce81b [rllib] Raise an error if multi-agent envs terminate without a last observation for agents (#4139)
* fix it

* lint

* Update rllib-training.rst
2019-02-23 21:23:40 -08:00
Philipp MoritzandRobert Nishihara ba52caff37 Make Bazel the default build system (#3898) 2019-02-23 11:58:59 -08:00
Tianming XuandRobert Nishihara 692bb336a1 Fix master branch compilation error and lint error (#4109) 2019-02-21 11:54:30 -08:00
Eric LiangandGitHub f8bef004da [rllib] Improve error message for bad envs, add remote env docs (#4044)
* commit

* fix up rew
2019-02-18 01:28:19 -08:00
Megan KawakamiandEric Liang 346885068c [rllib] add torch pg (#3857)
* add torch pg

* add torch imports

* added torch pg

* working torch pg implementation

* add pg pytorch

* Update a3c.py

* Update a3c.py

* Update torch_policy_graph.py

* Update torch_policy_graph.py
2019-02-16 19:54:14 -08:00
Hao ChenandGitHub de17443dc2 Propagate backend error to worker (#4039) 2019-02-16 11:39:15 +08:00
Robert NishiharaandPhilipp Moritz 5f71751891 API cleanups. Remove worker argument. Remove some deprecated arguments. (#4025)
* Remove worker argument from API methods.

* Remove deprecated arguments and deprecate redirect_output and redirect_worker_output.

* Fix
2019-02-15 10:49:16 -08:00
William MaandRobert Nishihara 8ee53297b1 Add documentation on how to use debug tools (#4000) 2019-02-14 13:50:21 -08:00
Philipp MoritzandRobert Nishihara 077ffd99bf Bump version from 0.6.3 to 0.7.0.dev0 in docs and .yaml (#4042) 2019-02-14 12:08:48 -08:00
Eric LiangandGitHub 2dccf383dd [rllib] Basic infrastructure for off-policy estimation (IS, WIS) (#3941) 2019-02-13 16:25:05 -08:00
Hao ChenandGitHub f31a79f3f7 Implement actor checkpointing (#3839)
* Implement Actor checkpointing

* docs

* fix

* fix

* fix

* move restore-from-checkpoint to HandleActorStateTransition

* Revert "move restore-from-checkpoint to HandleActorStateTransition"

This reverts commit 9aa4447c1e3e321f42a1d895d72f17098b72de12.

* resubmit waiting tasks when actor frontier restored

* add doc about num_actor_checkpoints_to_keep=1

* add num_actor_checkpoints_to_keep to Cython

* add checkpoint_expired api

* check if actor class is abstract

* change checkpoint_ids to long string

* implement java

* Refactor to delay actor creation publish until checkpoint is resumed

* debug, lint

* Erase from checkpoints to restore if task fails

* fix lint

* update comments

* avoid duplicated actor notification log

* fix unintended change

* add actor_id to checkpoint_expired

* small java updates

* make checkpoint info per actor

* lint

* Remove logging

* Remove old actor checkpointing Python code, move new checkpointing code to FunctionActionManager

* Replace old actor checkpointing tests

* Fix test and lint

* address comments

* consolidate kill_actor

* Remove __ray_checkpoint__

* fix non-ascii char

* Loosen test checks

* fix java

* fix sphinx-build
2019-02-13 19:39:02 +08:00
Si-YuanandRobert Nishihara 21472b890a Integrate "tempfile_service" into "ray.node.Node" (#3953) 2019-02-12 17:34:04 -08:00
Adi ZimmermanandRichard Liaw dac1969647 [tune] Add Nevergrad to Tune (#3985) 2019-02-12 11:00:04 -08:00
Adi ZimmermanandRichard Liaw 9797028a91 [tune] Add scikit-optimize to Tune (#3924) 2019-02-11 17:06:02 -08:00
Eric LiangandGitHub c4182463f6 [rllib] Add helper to iterate over envs in a vectorized environment (#4001)
* add foreach env func

* fix

* add test
2019-02-11 10:40:47 -08:00
Robert NishiharaandPeter Schafhalter 6a32b410bb Update versions from 0.6.2 -> 0.6.3 in the documentation. (#3981) 2019-02-07 20:57:37 -08:00
Alex LaGrassaandRobert Nishihara b0fe5af7c8 [doc] Update example-parameter-server.rst (#3773) 2019-02-05 22:00:54 -08:00
Andrew TanandRichard Liaw 8323419a6d [tune] Add SigOpt Integration (#3844) 2019-02-03 18:23:57 -08:00
Michael LuoandEric Liang 1a015e420b Optimal PPO Configs (10k reward in 1 hr) + PPO grad clipping implemented (#3934) 2019-02-02 22:10:58 -08:00
Peter SchafhalterandRichard Liaw 62a0a7bdc7 [tune] Add BayesOpt (#3864)
Adds BayesOpt as a Tune suggestion algorithm.
2019-01-31 16:54:17 -08:00
Philipp MoritzandRichard Liaw beb75193da Fix linting on master (#3913) 2019-01-31 01:28:45 -08:00
Rong OuandRobert Nishihara 8f6bd6cece change kubernetes examples to use Deployment (#3909) 2019-01-30 17:50:37 -08:00
Eric LiangandGitHub 152375aa8a [rllib] Add evaluation option to DQN agent (#3835)
* add eval

* interval

* multiagent minor fix

* Update rllib.rst

* Update ddpg.py

* Update qmix.py
2019-01-29 21:19:53 -08:00
Eric LiangandGitHub fb73cedf70 [rllib] Add examples page, add hierarchical training example, delete SC2 examples (#3815)
* wip

* lint

* wip

* up

* wip

* update examples

* wip

* remove carla

* update

* improve envspec

* link to custom

* Update rllib-env.rst

* update

* fix

* fn

* lint

* ds

* ssd games

* desc

* fix up docs

* fix
2019-01-29 21:06:09 -08:00
Stephanie WangandHao Chen eddd60e14e Improve backend debug logging, refactor scheduling queues (#3819) 2019-01-26 16:15:48 +08:00
Si-YuanandPhilipp Moritz 48139cf861 Migrate Python C extension to Cython (#3541) 2019-01-24 09:17:14 -08:00
Eric LiangandGitHub 04ec47cbd4 [rllib] annotate public vs developer vs private APIs (#3808) 2019-01-23 21:27:26 -08:00
Robert NishiharaandPhilipp Moritz 01e18b47f4 Direct people to stackoverflow for questions about usage. (#3830)
* Direct people to stackoverflow for questions about usage.

* Improve wording
2019-01-23 13:30:02 -08:00
Robert NishiharaandPhilipp Moritz 0b1608a546 Factor out code for starting new processes and test plasma store in valgrind. (#3824)
* Factor out starting Ray processes.

* Detect flags through environment variables.

* Return ProcessInfo from start_ray_process.

* Print valgrind errors at exit.

* Test valgrind in travis.

* Some valgrind fixes.

* Undo raylet monitor change.

* Only test plasma store in valgrind.
2019-01-22 14:59:11 -08:00
Michael LuoandEric Liang 16f7ca45e4 Appo (#3779)
* Deleted old fork, updated new ray and moved PPO-impala to APPO in ppo folder

* Deleted unneccesary vtrace.py file

* Update pong-impala.yaml

* Cleaned PPO Code

* Update pong-impala.yaml

* Update pong-impala.yaml

* wip

* new ifle

* refactor

* add vtrace off option

* revert

* support any space

* docs

* fix comment

* remove kl

* Update cartpole-appo-vtrace.yaml
2019-01-18 13:40:26 -08:00
Richard LiawandRobert Nishihara 0537508106 Bump strings for 0.6.2 (#3801) 2019-01-17 19:03:27 -08:00
Jones WongandEric Liang 319c1340cb [rllib] Develop MARWIL (#3635)
*  add marvil policy graph

*  fix typo

*  add offline optimizer and enable running marwil

*  fix loss function

*  add maintaining the moving average of advantage norm

*  use sync replay optimizer for unifying

*  remove offline optimizer and use sync replay optimizer

*  format by yapf

*  add imitation learning objective

*  fix according to eric's review

*  format by yapf

* revise

* add test data

* marwil
2019-01-16 19:00:43 -08:00
Eric LiangandRichard Liaw 401e656b95 [rllib] Sync filters at end of iteration not start; hierarchical docs (#3769) 2019-01-15 16:25:25 -08:00
jhpengerandRichard Liaw 3adffe6a4e [docs] Add example showing how to use Ray on Kubernetes. (#3126)
Closes #1353.
2019-01-13 13:56:47 -08:00
Robert NishiharaandRichard Liaw 1480f309c3 [doc] Replace runtest.py with mini_test.py in documentation. (#3750)
Rename `xray_test.py` to `mini_test.py` and use that in the documentation. Right now we suggest that people run `runtest.py`, but that often doesn't succeed and takes too long.
2019-01-12 14:05:28 -08:00
Eric LiangandGitHub e78562b2e8 [rllib] Misc fixes: set lr for PG, better error message for LSTM/PPO, fix multi-agent/APEX (#3697)
* fix

* update test

* better error

* compute

* eps fix

* add get_policy() api

* Update agent.py

* better err msg

* fix

* pass in rew
2019-01-06 19:37:35 -08:00
Eric LiangandGitHub 03fe760616 [rllib] Model self loss isn't included in all algorithms (#3679) 2019-01-04 22:30:35 -08:00
Eric LiangandRichard Liaw 7db1f3be2a [tune] resume=False by default but print a tip to set resume="prompt" + jenkins fix (#3681) 2019-01-04 17:23:19 -08:00
Robert NishiharaandPhilipp Moritz 586a5c9ffa Limit default redis max memory to 10GB. (#3630)
* Limit Redis max memory to 10GB/shard by default.

* Update stress tests.

* Reorganize

* Update

* Add minimum cap size for object store and redis.

* Small test update.
2019-01-03 13:23:54 -08:00
Eric LiangandGitHub ca864faece [rllib] Documentation for I/O API and multi-agent support / cleanup (#3650) 2019-01-03 15:15:36 +08:00
Eric LiangandGitHub 47d36d7bd6 [rllib] Refactor pytorch custom model support (#3634) 2019-01-03 13:48:33 +08:00
Robert NishiharaandPhilipp Moritz b6bcd18d65 Split profile table among many keys in the GCS. (#3676)
* Divide profile table among many keys in GCS.

* Fix, and remove --collect-profiling-data arg.

* Remove reference in doc.
2019-01-02 21:33:01 -08:00
Eric LiangandGitHub b8a9e3f106 [rllib] Remove uses of sgd_stepsize => lr (#3667)
* lr

* Update example-evolution-strategies.rst
2019-01-01 12:01:27 +08:00
Richard LiawandGitHub aad3c50e2d [tune] Cluster Fault Tolerance (#3309)
This PR introduces cluster-level fault tolerance for Tune by checkpointing global state. This occurs with relatively high frequency and allows users to easily resume experiments when the cluster crashes.

Note that this PR may affect automated workflows due to auto-prompting, but this is resolvable.
2018-12-29 11:42:25 +08:00
Robert NishiharaandPhilipp Moritz 5426234cd8 Update documentation to reflect 0.6.1 release. (#3622) 2018-12-24 11:10:04 -08:00
Eric LiangandGitHub 9f63119a83 [rllib] Allow development without needing to compile Ray (#3623)
* wip

* lint

* wip

* wip

* rename

* wip

* Cleaner handling of cli prompt
2018-12-24 18:08:23 +09:00
Si-YuanandRobert Nishihara a1995ff3b0 Resize logo in README. (#3619) 2018-12-23 22:59:23 -08:00