wassname/ray
1
0
Fork 0
mirror of https://github.com/wassname/ray.git synced 2026-07-07 14:24:44 +08:00
Code Issues Packages Projects Releases Wiki Activity
2,999 Commits 36 Branches 42 Tags
dfc94ce7bcd5d9d008822efdeec17c3f6bb9c606
Commit Graph

2999 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Stefan Pantic dfc94ce7bc [rllib]Add entropy coeff decay (#5043) 2019-07-08 18:30:32 -07:00
Daniel Edgecumbe eeb67db861 [autoscaler] Log AWS NodeProvider create_instances (#4998) 
* autoscaler: Log on AWS NodeProvider create_instances

* logging
 2019-07-08 13:22:26 -07:00
Hao Chen 8a30b93e42 Define common data structures with protobuf. (#5121) 2019-07-08 22:41:37 +08:00
Joey Jiang b4e51c8aa1 Support clang-format whose version is not 7.0 (#5139) 2019-07-08 17:15:09 +08:00
Sam Toyer 7ad854d4c6 [tune] Use traceback.format_tb() (fixes #5135) (#5136) 2019-07-08 01:13:06 -07:00
Joey Jiang 274233962f Remove unused connection file in object manager (#5123) 2019-07-08 10:59:36 +08:00
Eric Liang 893744b3be [rllib] Revert "use make template" which seems to break DQN/Atari (#5134) 
* Revert "use make template"

This reverts commit 291e9e0031c6e315fe24e5b4973dea375fe73918.

* debug vars
 2019-07-07 19:51:26 -07:00
Morgan Giraud 7e020e7183 [tune] tune.run keep_checkpoints_num (#5117) 
* Add missing argument keep_checkpoints_num to tune

* expose keep checkpoints
 2019-07-07 17:14:56 -07:00
Edward Oakes 8f53364097 Improve local_mode (#5060) 2019-07-07 17:10:50 -07:00
Eric Liang 932d6b2517 [rllib] Port IMPALA to ModelV2/build_tf_policy (#5130) 
* port vtrace

* fix vf

* fix vs

* fix the example

* wip ddpg

* fix tests

* fix tests

* remove ddpg model

* comments

* set vf share layers True by default

* typo

* fix test
 2019-07-07 15:06:41 -07:00
Richard Liaw 6a14f1a540 [autoscaler] Small fixes for local cluster usability (#4864) 2019-07-06 21:55:18 -07:00
Richard Liaw 1798d4f077 [autoscaler] Add hard kill and monitor commands (#5082) 
* Add hard kill and monitor commands

* better_commands

* Update python/ray/scripts/scripts.py

Co-Authored-By: Kristian Hartikainen <kristian.hartikainen@gmail.com>
 2019-07-06 21:52:55 -07:00
Eric Liang 445bcb29b0 [hotfix] fix backward compat with older yaml libraries 2019-07-06 20:41:28 -07:00
Eric Liang c15ed3ac55 [rllib] Shuffle RNN sequences in PPO as well (#5129) 
* shuffle seq

* fix test
 2019-07-06 20:40:49 -07:00
Brandon Bertelsen c04b69902c Updates for #5072 (#5091) 2019-07-06 16:05:50 -07:00
Eric Liang 0448847a02 Update protobuf version (#5128) 2019-07-06 15:59:55 -07:00
Aleksei Petrenko 09bde397c9 Multiagent experiment resume (#5102) 
* Fixed problem with multiagent experiment resume

* Applied format script

* fix lint
 2019-07-06 11:38:17 -07:00
Dušan Josipović e9b88dcbed [wingman -> tune] Add system performance tracking (#4924) 2019-07-06 00:57:35 -07:00
Richard Liaw c3e9d94b18 [tune][minor] Reduce checkpointing frequency (#4859) 2019-07-06 00:54:24 -07:00
Kim Jeong Ju 4b56a5eb27 [tune] missing torch.load in mnist_pytorch_trainable.py (#5103) 2019-07-06 00:14:41 -07:00
Philipp Moritz c5253cc300 Add job table to state API (#5076) 2019-07-06 00:05:48 -07:00
Richard Liaw 53d5a8a45f [tune] Fix sort (#5111) 
* fix sort

* fix tune list-experiments

* Update python/ray/tune/tests/test_commands.py
 2019-07-05 16:05:10 -07:00
Joey Jiang 4183303a2f Add bazel build options for plasma to use glog (#5108) 2019-07-05 19:00:19 +08:00
Robert Nishihara 9cc4cc6a52 Fail format.sh if yapf/flake8 versions are incorrect. (#5083) 2019-07-04 23:22:01 -07:00
Zhijun Fu 54d5969cea [grpc] Add grpc server to worker (#5054) 
* refactor grpc server

* format

* change GetTask() to PushTask()

* change PushTask to AssignTask

* format

* update

* fix test

* format

* Update src/ray/rpc/worker_client.h

Co-Authored-By: Hao Chen <chenh1024@gmail.com>

* Update BUILD.bazel

* Update src/ray/core_worker/task_execution.cc

Co-Authored-By: Stephanie Wang <swang@cs.berkeley.edu>

* update

* format

* address comments

* format

* Update src/ray/rpc/worker/worker_server.h

Co-Authored-By: Stephanie Wang <swang@cs.berkeley.edu>

* Update src/ray/protobuf/worker.proto

Co-Authored-By: Stephanie Wang <swang@cs.berkeley.edu>

* format

* fix

* format
 2019-07-04 20:16:42 +08:00
ztangent 41a16c55ef [tune] Fixed bug with joining experiment_path twice. (#5106) 2019-07-03 22:48:07 -07:00
Patrick 1a543a6571 [serve] add missing __init__.py file under serve/utils (#4609) 
* bugfix: add missing serve/utils __init__.py file

* Update __init__.py

* lint
 2019-07-03 17:27:59 -07:00
Richard Liaw 0dbb6c4911 [tune] PBT perturbing after first iteration (#5097) 2019-07-03 17:27:26 -07:00
Eric Liang 34d054ff19 [rllib] ModelV2 API (#4926) 2019-07-03 15:59:47 -07:00
Kristian Hartikainen 9e0192bc0b [tune] Change the log syncing behavior (#4450) 
* Change the log syncing behavior

* fix up abstractions for syncer

* Finished checkpoint syncing

* Code

* Set of changes to get things running

* Fixes for log syncing

* Fix parts

* Lint and other fixes

* fix some test

* Remove extra parsing functionality

* some test fixes

* Fix up cloud syncing

* Another thing to do

* Fix up tests and local sync

Changes LogSync into a mixin, and adds tests for different
functionalities.

* Fix up tests, start on local migration

* fix distributed migrations

* comments

* formatting

* Better checkpoint directory handling

* fix tests

* fix tests

* fix click

* comments

* formatting comments

* formatting and comments

* sync function deprecations

* syncfunction

* Add documentation for Syncing and Uploading

* nit

* BaseSyncer as base for Mixin in edge case

* more docs

* clean up assertions

* validate

* nit

* Update test_cluster.py

* betterdoc

* Update tune-usage.rst

* cleanup

* nit
 2019-07-02 20:46:00 -07:00
Stephanie Wang 71d4637b75 [core worker] Refactor CoreWorker member classes (#5062) 
* Move store client mutex inside CoreWorkerPlasmaStoreProvider

* Move PlasmaClient inside CoreWorkerStoreProvider

* Remove CoreWorkerObjectInterface's ref to CoreWorker

* Remove WorkerLanguage

* Remove CoreWorkerTaskInterface's ref to CoreWorker

* Remove CoreWorkerTaskExecutionInterface's ref to CoreWorker

* lint

* move comment

* Fix build

* Fix build
 2019-07-02 15:30:30 -07:00
Kai Yang 1cf7728f35 [Core worker] Serialize ActorHandle in core worker. Make ActorHandle thread safe. (#5034) 
* Serialize ActorHandle in core worker. Make ActorHandle thread safe.

* Address comments

* Address comments

* Address comments

* Address comments

* lint

* Address comments

* Address comments

* Address comments

* Address comments

* Minor update

* Address comments

* lint
 2019-07-02 16:48:43 +08:00
Eric Liang 904dcf081d Switch cluster longevity tests to DLAMI, fix ray up verbosity (#5084) 
* fix

* add branch commit

* comments

* Update ci/long_running_tests/.gitignore

Co-Authored-By: Robert Nishihara <robertnishihara@gmail.com>
 2019-07-02 00:19:05 -07:00
Qing Wang 247f95b3ff Refine RegisterClientRequest message to make it clearer. (#5057) 
* transfor driver task id Explicitly

* Refins

* Fix and add comment.

* add more

* Fix

* Fix

* Add comments

* Fix
 2019-07-02 14:26:19 +08:00
Philipp Moritz a6a02fccd0 Do not compile redis twice (#5074) 2019-07-01 15:42:54 -07:00
Philipp Moritz 4e82313891 Update to latest arrow (#5011) 2019-06-30 20:36:36 -07:00
Simon Mo 0c4dd3c401 Use bazel disk cache with travis (#5068) 2019-06-30 17:57:48 -07:00
Simon Mo 6c4c1d444d Update VersionKey in stats (#5070) 2019-06-30 18:23:12 +08:00
Simon Mo d7ccfbe46b Bump version to 0.8.0.dev2 (#5069) 2019-06-29 23:30:26 -07:00
Simon Mo b5d473847c bump version to 0.7.2 (#5066) 2019-06-29 19:06:51 -07:00
Robert Nishihara bcc379556b Make some fixes to long running stress tests. (#5056) 2019-06-28 15:42:54 -07:00
Kai Yang 4ccb7b05cc [Core worker] Add metadata support in object interface (#5031) 2019-06-28 11:35:03 -07:00
Hao Chen cefbb0c94c Fix driver id in TaskInfo (#5055) 2019-06-28 12:56:48 +08:00
Kai Yang a39982e676 [Core worker] Task execution passes TaskInfo struct to executor (#5032) 2019-06-28 10:59:45 +08:00
Joey Jiang d6bbbdef35 Use gRPC to handle communication and data transmission between object manager (#4996) 2019-06-28 10:56:34 +08:00
Qing Wang 62e4b591e3 [ID Refactor] Rename DriverID to JobID (#5004) 
* WIP

WIP

WIP

Rename Driver -> Job

Fix complition

Fix

Rename in Java

In py

WIP

Fix

WIP

Fix

Fix test

Fix

Fix C++ linting

Fix

* Update java/runtime/src/main/java/org/ray/runtime/config/RayConfig.java

Co-Authored-By: Stephanie Wang <swang@cs.berkeley.edu>

* Update src/ray/core_worker/core_worker.cc

Co-Authored-By: Stephanie Wang <swang@cs.berkeley.edu>

* Address comments

* Fix

* Fix CI

* Fix cpp linting

* Fix py lint

* FIx

* Address comments and fix

* Address comments

* Address

* Fix import_threading
 2019-06-28 00:44:51 +08:00
Qing Wang d9768c1cd2 [hotfix] Fix master's linting (#5049) 
The linting in CI on master always fail.
 2019-06-27 20:21:32 +08:00
Hao Chen a1156754e9 Fix test_task_forward (#5040) 2019-06-27 14:37:00 +08:00
Hao Chen 469ae41013 Fix memory leak in rpc ServerCall and ClientCall (#5046) 2019-06-27 13:19:47 +08:00
Daniel Edgecumbe 49c6e81de2 autoscaler/monitor: Kill workers on exception (#4997) 2019-06-26 17:59:12 -07:00