Commit graph

8760 commits

Author SHA1 Message Date
Jiao
994ff3ce21
[Serve] Add initial large scale tests (#17026) 2021-07-20 08:56:29 -07:00
Kai Yang
f0c148b158
[Core] Simplify the code to read env variables in RayConfig (#16775)
* Simplify the code to read env variables in RayConfig

* simplify

* Correctly print config type

* Change to lower case

* fix template specialization

* lint
2021-07-20 08:40:16 -07:00
SangBin Cho
d6b6356173
[Core] Properly call shutdown instead of deleting a reference (#17096)
* Properly call shutdown instead of deleting a reference

* Add unit tests

* Add test ray shutdown

* Formatting

* format2

* Revert main logic to see if windows issue still fail

* Skip tests for windows.

* formatting

* Try fixing flakiness

* Remove node removed code path

Co-authored-by: Eric Liang <ekhliang@gmail.com>
2021-07-20 08:22:33 -07:00
Antoni Baum
5e9b680e39
[docs] Add LightGBM-Ray docs, update XGBoost-Ray docs (#17188) 2021-07-20 16:06:47 +01:00
Siyuan (Ryans) Zhuang
8efc04a8a6
[Core] Actor namespace (#17178)
* set actor namespace in Python on creation

* get actor with namespace in Python

* update message
2021-07-19 21:51:04 -07:00
matthewdeng
fef74aa94f
[sgd] add placement group support (#17037)
* [sgd] add placement group support

* add logic for removing placement group upon shutdown

* set placement group; add tests

* address comments - add timeout and improve error handling

* remove unused import

* mock SGD_PLACEMENT_GROUP_TIMEOUT_S
2021-07-19 21:50:37 -07:00
Siyuan (Ryans) Zhuang
9ca6bda3a1
[Workflow] Fix recovery storage mismatch issue (#17166)
* fix recovery path issue and add test

* add TODOs
2021-07-19 21:49:12 -07:00
dependabot[bot]
2de7b8f084
[tune](deps): Bump tune-sklearn in /python/requirements/tune (#17173)
Bumps [tune-sklearn](https://github.com/ray-project/tune-sklearn) from 0.3.0 to 0.4.0.
- [Release notes](https://github.com/ray-project/tune-sklearn/releases)
- [Commits](https://github.com/ray-project/tune-sklearn/compare/v0.3.0...v0.4.0)

---
updated-dependencies:
- dependency-name: tune-sklearn
  dependency-type: direct:production
  update-type: version-update:semver-minor
...

Signed-off-by: dependabot[bot] <support@github.com>

Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2021-07-19 19:42:54 -07:00
Amog Kamsetty
c9522e9a6f
Remove requests from Core Dependencies (#17066)
* remove requests

* update

Signed-off-by: Richard Liaw <rliaw@berkeley.edu>

* update

* lint

Co-authored-by: Richard Liaw <rliaw@berkeley.edu>
2021-07-19 19:38:29 -07:00
Eric Liang
fabba96fad
Re-merge large function def, skipping test failing on Windows (#17191) 2021-07-19 18:03:26 -07:00
Chen Shen
fe9a6b669c
[nightly-test] add 4-nodes shuffle-data-loader test (#17155) 2021-07-19 17:46:22 -07:00
Chen Shen
b26fcd3fce
fix spill bug (#17187) 2021-07-19 17:44:12 -07:00
Amog Kamsetty
777921b2e7
[dependencies] vendor colorama (#17183) 2021-07-19 16:29:29 -07:00
Eric Liang
d59da075a6
Re-merge TMPDIR support, but only for Linux. OSX requires RAY_TMPDIR (#17190) 2021-07-19 15:45:03 -07:00
SangBin Cho
561dcbd99c
[Test] Fix the permission issue for Dask on Ray multi node sort #17189( #17189) 2021-07-19 14:42:39 -07:00
Patrick Ames
34789b3e56
[autoscaler] Add support for custom EC2 instance network interfaces (#14080)
* [autoscaler] Add support for custom EC2 instance network interfaces.

* [autoscaler] Add unit tests for custom EC2 network interfaces and support for AWS node provider stubs.
2021-07-19 17:21:21 -04:00
Chen Shen
80e013f342
[core] Fix SIGABRT on erase call (#17140) 2021-07-19 11:42:38 -07:00
SangBin Cho
bfc9e5c36f
[Logs] Clean core worker logs (#17033)
* Ready

* Formatting

* Fix

* addressed review.
2021-07-19 11:25:41 -07:00
Amog Kamsetty
cb74053ee5
Retry remove gpustat dependency (#17115)
* remove gpustat

* move psutil imports
2021-07-19 11:14:10 -07:00
Siyuan (Ryans) Zhuang
1fbfbfc55b
[Serializatioin] Bump pickle5 version (#17124) 2021-07-19 10:40:38 -07:00
Siyuan (Ryans) Zhuang
9b110f9344
[Workflow] Update API (#17165)
* actor_id same as workflow_id

* @workflow.actor -> @workflow.virtual_actor

* readonly decorator

* run/run_async for virtual actor

* get_or_create for virtual actor

* update doc

* run/run_async for steps

* update tests

* update comments
2021-07-19 10:19:46 -07:00
Sven Mika
18d173b172
[RLlib] Implement policy_maps (multi-agent case) in RolloutWorkers as LRU caches. (#17031) 2021-07-19 13:16:03 -04:00
Sven Mika
e0640ad0dc
[RLlib] Fix seeding for ES and ARS. (#16744) 2021-07-19 13:13:05 -04:00
architkulkarni
4069686e0f
Revert "Improve error message for oversized function (#17133)" (#17184)
This reverts commit 3e53619d64.
2021-07-19 09:28:33 -07:00
Qing Wang
195cdcf5b8
Fix memory leak in JNI. (#17177)
Co-authored-by: Qing Wang <jovany.wq@antgroup.com>
2021-07-19 14:06:30 +08:00
Amog Kamsetty
8dfd471823
Revert "Revert "[Dashboard][event] Basic event module (#16985)" (#17068)" (#17107)
This reverts commit c17e171f92.

Co-authored-by: 刘宝 <po.lb@antfin.com>
2021-07-18 12:59:04 +08:00
dependabot[bot]
bda1a37e93
[tune](deps): Bump mlflow in /python/requirements/tune (#17168)
Bumps [mlflow](https://github.com/mlflow/mlflow) from 1.17.0 to 1.19.0.
- [Release notes](https://github.com/mlflow/mlflow/releases)
- [Changelog](https://github.com/mlflow/mlflow/blob/master/CHANGELOG.rst)
- [Commits](https://github.com/mlflow/mlflow/compare/v1.17.0...v1.19.0)

---
updated-dependencies:
- dependency-name: mlflow
  dependency-type: direct:production
  update-type: version-update:semver-minor
...

Signed-off-by: dependabot[bot] <support@github.com>

Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2021-07-17 16:23:10 -07:00
Eric Liang
3e53619d64
Improve error message for oversized function (#17133) 2021-07-17 11:04:05 -07:00
Philipp Moritz
c5c167300b
[Ray debugger] Set breakpoint() hook only for tasks and actors (#17164) 2021-07-17 10:27:51 -07:00
Simon Mo
9da49a7fb8
Revert "ray ignores TMPDIR variable" (#17161)
This reverts commit c27f43d9b8.
2021-07-16 20:28:38 -07:00
Eric Liang
94f17ec099
[RFC] API stability annotations (#17100) 2021-07-16 17:09:20 -07:00
Alex Wu
93c16346bf
[Dataset] imagenet nightly test (#17069) 2021-07-16 14:15:49 -07:00
Eric Liang
c27f43d9b8
ray ignores TMPDIR variable (#17130) 2021-07-16 13:23:44 -07:00
Eric Liang
26a286655b
Add link to datasets preview docs 2021-07-16 12:31:52 -07:00
Simon Mo
e7ede45e37
[Buildkite] Cleanup macOS builders (#17145)
macOS builders are reused, so excessive disk usage might lead to
run out of disk space error
2021-07-16 10:46:08 -07:00
Clark Zinzow
8302b5a335
[Core] Reverts full dispatch queue iteration PRs. (#17127)
* Revert "[Core] iterate over entire dispatch queue instead of returning when worker unavailable (#16535)"

This reverts commit 54d66ac637.

* Revert "[Core] [runtime env] [Tests] Add C++ unit test for dispatch queue nonblocking behavior (#16751)"

This reverts commit 13a133817b.

* Revert failing runtime_env test.
2021-07-16 10:28:00 -07:00
SangBin Cho
246f80961e
Dask on Ray version documentation update (#16905)
* In progress

* done

* Fix the table format

* completed

* done

* Fix lint
2021-07-16 10:10:26 -07:00
Siyuan (Ryans) Zhuang
fd3742bb63
[Workflow] Update tests (#17147)
* update workflow tests

* use conftest

* use conftest

* use conftest
2021-07-16 09:25:40 -07:00
Eric Liang
40f1ee6e1b
Fix prefetch typo in Dataset (#17143) 2021-07-16 09:22:42 -07:00
fyrestone
e2808a35cf
Dashboard job module uses attrs instead of pydantic for job description (#17116) 2021-07-16 22:26:00 +08:00
SongGuyang
21b464ae9d
[C++ API] support get ray address from env (#17144) 2021-07-16 17:17:43 +08:00
Kai Fricke
49b72eec16
[tune] filter placement group resources if not in use (#16996) 2021-07-16 00:35:04 -07:00
Zhi Lin
6d9fb421c6
[tracing] Do not wrap when tracing is not enabled (#16607) 2021-07-16 00:27:54 -07:00
SangBin Cho
ef1d9278b8
[Test] nightly test dask on ray multi node sort (#17141) 2021-07-15 23:13:35 -07:00
Eric Liang
f03b43c532
[dataset] Support callable classes to simplify state initialization (#17136) 2021-07-15 23:06:14 -07:00
SongGuyang
dcb1baabd7
[C++ API] support loading C++ dynamic libraries from code search path (#16828) 2021-07-16 13:02:45 +08:00
SongGuyang
a57de0e224
support build different python wheel in setup.py (#16998) 2021-07-16 13:01:48 +08:00
Eric Liang
e69987bc96
Improve dataset error message (#17129) 2021-07-15 20:58:15 -07:00
Edward Oakes
90a1667b29
[debugger] Clean up breakpoint state for dead jobs (#17095) 2021-07-15 22:20:09 -05:00
Chen Shen
2a53d22438
[nightly-test] add test shuffle_data_loader (#16972)
* test shuffle_data_loader

* address comments

* update
2021-07-15 20:03:35 -07:00