wassname/ray
1
0
Fork 0
mirror of https://github.com/wassname/ray.git synced 2026-06-29 20:17:17 +08:00
Code Issues Packages Projects Releases Wiki Activity
6,812 Commits 36 Branches 42 Tags
python3.9
Commit Graph

6812 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Eric Liang f48da50e1c [rllib] observation function api for multi-agent (#8236) 2020-05-04 22:13:49 -07:00
Simon Mo 1480bf4295 [Serve] Improve batch size inconsistency error (#8315) 2020-05-04 20:32:12 -07:00
Simon Mo ca929671b6 [Serve] Simplify Validation (#8316) 2020-05-04 20:31:23 -07:00
Rüdiger Busche e93ec3134a Use kubectl delete pod in example (#8295) 
Co-authored-by: rbusche <rbusche@inserve.de>
 2020-05-04 21:39:30 -05:00
Rüdiger Busche 5dd9dbf74f Add ipython as dependency for autoscaler container (#8297) 
Co-authored-by: rbusche <rbusche@inserve.de>
 2020-05-04 21:22:38 -05:00
fangfengbin 14d03a0869 GCS adapts to task lease table pub sub (#8299) 2020-05-05 10:16:56 +08:00
ijrsvt cc7bd6650a [core] Enabling Remote Task Cancelation (#8225) 2020-05-04 15:24:22 -07:00
Sven Mika 6c2b9a4cfa [RLlib] Remove tf.py_function from all Schedule classes (not differentiable and causes other bugs in MA setups). (#8304) 
Remove tf.py_function from all Schedule classes (not differentiable and causes other bugs in MA setups). (#8304)
 2020-05-04 23:53:38 +02:00
Sven Mika a00144f746 [RLlib] Fix issue 8135 (DDPG inf actions when using [-inf,inf] action space). (#8302) 2020-05-04 22:27:30 +02:00
Stephanie Wang 8625e09067 Actor manager refactor and unit tests (#8224) 
* parametrize test

* Regression test and logging

* Test no restart after actor deletion

* Unit tests

* Refactor to subscribe to and lookup from worker failure table

* Refactor ActorManager to remove dependencies

* Revert "Regression test and logging"

This reverts commit 835e1a9091b51ca8efb00392d4cc4a665145de24.

* Revert "parametrize test"

This reverts commit f31272082831ba1a494816dd5511d87b24eca4c9.

* Revert "Test no restart after actor deletion"

This reverts commit 114a83de14329aa6ab787c80cd5757cf074a9072.

* doc

* merge

* Revert "Refactor to subscribe to and lookup from worker failure table"

This reverts commit 6aa13a05178d0b9aa1db9dee5c978c911b74fa3a.

* Use actor ID instead of shared_ptr

* TODO and lint

* Update src/ray/gcs/gcs_server/gcs_actor_scheduler.h

Co-authored-by: Edward Oakes <ed.nmi.oakes@gmail.com>

* Fix build

* doc

* Build

Co-authored-by: Edward Oakes <ed.nmi.oakes@gmail.com>
 2020-05-04 10:16:52 -07:00
Sven Mika b95e28faea [RLlib] APEX_DDPG (PyTorch) test case and docs. (#8288) 
APEX_DDPG (PyTorch) test case and docs.
 2020-05-04 09:36:27 +02:00
fangfengbin 5f351a05fe GCS adapts to task table pub sub (#8210) 2020-05-04 10:23:55 +08:00
Ion e24276e3c1 New scheduler int capacities (#8192) 
* working version

* working version

* done

* done

* done

* addressing most of Philipp comments

* addressing most of Philipp comments
 2020-05-03 18:47:30 -07:00
Sven Mika 2c71559b3a [RLlib] Fixes duplicate learning tests in travis.yaml. (#8293) 
Fixes duplicate learning tests in travis.yaml.
 2020-05-03 20:27:29 +02:00
fangfengbin b7bbc3bc83 [GCS]GCS adapts to object table pub sub (#8180) 2020-05-03 21:44:33 +08:00
Sven Mika 166bb5d690 [RLlib] IMPALA PyTorch (#8287) 
This PR adds an IMPALA PyTorch implementation.

- adds compilation tests for LSTM and w/o LSTM.
- adds learning test for CartPole.
 2020-05-03 13:44:25 +02:00
Eric Liang 1228369a87 Remove "This tab is experimental" (#8281) 2020-05-02 22:41:28 -07:00
Simon Mo ec6631ae58 Pin redis-py version (#8290) 2020-05-02 22:09:02 -07:00
Stephanie Wang b4ef500675 [core] Disable GCS actor management (#8271) 2020-05-02 22:02:56 -07:00
SangBin Cho 0f54d5ab65 Async actor microbenchmark Script (#8275) 2020-05-02 21:51:00 -07:00
Richard Liaw 40dfb337bf [tune] Hotfix Ax breakage when fixing backwards-compat (#8285) 2020-05-02 20:42:50 -07:00
Xianyang Liu eda526c154 [SGD] Support multiple input model (#8246) 2020-05-02 16:49:09 -07:00
Maksim Smolin c2acb7ffe2 [SGD] Add imagenet example CI (#8150) 2020-05-02 16:48:35 -07:00
Edward Oakes 518ef4c0b3 [serve] Increase timeout waiting for HTTP server (#8286) 2020-05-02 16:55:13 -05:00
mehrdadn ff68fb8c7c Try to fix tests running all the time (#8280) 
Co-authored-by: Mehrdad <noreply@github.com>
 2020-05-02 15:37:52 -05:00
Edward Oakes 22cab930cd Retry actor failures in serve failure test (#8282) 2020-05-02 10:19:44 -05:00
Edward Oakes 8d3236f1d0 Lower test_utils.wait_for_condition default timeout to 30s (#8283) 2020-05-02 10:19:00 -05:00
Sven Mika 76e1a4df9e Fix TD3 torch via GaussianNoise torch bug. (#8276) 2020-05-02 08:12:21 +02:00
Edward Oakes 019030cb4d Add long-running serve failure test (#8277) 2020-05-01 21:07:14 -05:00
Edward Oakes d4e64709ba Shorten test_joblib (#8273) 2020-05-01 17:11:32 -05:00
mehrdadn bf074073e7 Deploy Windows wheels to Amazon S3 (#8237) 
* Deploy to Amazon S3

* Install specifically requested Python version

Co-authored-by: Mehrdad <noreply@github.com>
 2020-05-01 14:08:57 -07:00
Sven Mika 42991d723f [RLlib] rllib/examples folder restructuring (#8250) 
Cleans up of the rllib/examples folder by moving all example Envs into rllibexamples/env (so they can be used by other scripts and tests as well).
 2020-05-01 22:59:34 +02:00
Edward Oakes 13f718846d [serve] Always use internal KV store (#8270) 2020-05-01 14:18:18 -05:00
Richard Liaw 07daff8794 [tune] Avoid breakage - soft deprecation warning for search algs (#8258) 2020-05-01 10:36:43 -07:00
Edward Oakes 3aec683f61 Avoid fate sharing with owner for detached actors (#8267) 2020-05-01 11:58:47 -05:00
Edward Oakes 63bc7dc522 service -> endpoint in router (#8269) 2020-05-01 11:55:34 -05:00
Edward Oakes 421b3c9d8b Fix serve long running test (#8268) 2020-05-01 11:54:27 -05:00
Eric Liang 2a0ad0b8ce [rllib] [hotfix] Remove assert that trips on pytorch multiagent (#8241) 2020-05-01 06:32:54 +02:00
Edward Oakes 6373c70661 [serve] Refactor BackendConfig (#8202) 2020-04-30 22:31:07 -05:00
Edward Oakes 95d187e556 [serve] Add delete_endpoint call (#8256) 2020-04-30 20:59:07 -05:00
Edward Oakes 484f68765c Fix resource_ids_ data race (#8253) 2020-04-30 18:55:54 -05:00
Edward Oakes 43be73e4cf [serve] Add delete_backend call (#8252) 2020-04-30 13:10:39 -05:00
Sven Mika c593fb09b7 [RLlib] Remove all f-strings to keep py3.5 compatibility. 2020-04-30 11:10:16 -07:00
Sven Mika eea75ac623 [RLlib] Beta distribution. (#8229) 2020-04-30 11:09:33 -07:00
Sven Mika b23b6addfc [RLlib] Stabilize Pendulum-v0 regression test cases. (#8232) 
Stabilize Pendulum regression test cases.
 2020-04-30 15:48:11 +02:00
Richard Liaw 05df80afad Extend timeout for test_tune_server (#8233) 2020-04-30 08:39:46 -05:00
Eric Liang baadbdf8d4 [rllib] Execute PPO using training workflow (#8206) 
* wip

* add kl

* kl

* works now

* doc update

* reorg

* add ddppo

* add stats

* fix fetch

* comment

* fix learner stat regression

* test fixes

* fix test
 2020-04-30 01:18:09 -07:00
Richard Liaw 35eac2671e [sgd] Resource limit lift for GPU test (#8238) 2020-04-30 00:24:48 -07:00
mehrdadn 254b1ec370 Set up testing and wheels for Windows on GitHub Actions (#8131) 
* Move some Java tests into ci.sh

* Move C++ worker tests into ci.sh

* Define run()

* Prepare to move Python tests into ci.sh

* Fix issues in install-dependencies.sh

* Reload environment for GitHub Actions

* Move wheels to ci.sh and fix related issues

* Don't bypass failures in install-ray.sh anymore

* Make CI a little quieter

* Move linting into ci.sh

* Add vitals test right after build

* Fix os.uname() unavailability on Windows

Co-authored-by: Mehrdad <noreply@github.com>
 2020-04-29 21:19:02 -07:00
Eric Liang ae54e0dc0a [rllib] Copy plasma memory before adding data to replay buffer 2020-04-29 14:17:54 -07:00