Commit graph

9388 commits

Author SHA1 Message Date
Sven Mika
ff7740e508
[RLlib] Fix the git-bisect helper/debugging script. (#18040) 2021-08-24 18:08:29 +02:00
Antoni Baum
1f8ce1ede8
[tune] Explicitly instantiate skopt categorical spaces (#18005) 2021-08-24 17:11:21 +02:00
Alex Wu
6e3dd7b3cf
Revert "[Core]make thread of client manager in gcs server configurable (#17978)" (#18041)
This reverts commit f0edbf0d30.
2021-08-24 07:57:59 -07:00
Qing Wang
7c1f14ddd8
Do not connect in constructor to avoid potential risk. (#17916)
* Do not connect in ctor.

* Fix lint.

Co-authored-by: Qing Wang <jovany.wq@antgroup.com>
2021-08-24 16:41:30 +08:00
Kai Fricke
ff68251f89
[release] Add python 3.9 to fix-docker-latest.sh (#18037) 2021-08-24 10:07:03 +02:00
Chen Shen
7e3e0d1535
[Test] Add C++ tsan test (#17875) 2021-08-24 00:57:32 -07:00
wanxing
abb46de4dc
[object store refactor 5/n] Add eviction policy tests (#17984)
* add eviction policy tests

* fix object_lifecycle_manager_test build

* make IsObjectExists private
2021-08-24 00:50:28 -07:00
Kai Fricke
d058f98546
[RLlib] Add GPU tests to CI (run per-PR). (#17891)
Co-authored-by: simon-mo <simon.mo@hey.com>
2021-08-24 09:20:45 +02:00
Tao Wang
f0edbf0d30
[Core]make thread of client manager in gcs server configurable (#17978) 2021-08-24 11:27:35 +08:00
dependabot[bot]
15adedc72c
[tune](deps): Bump sigopt in /python/requirements/tune (#17996)
Bumps [sigopt](https://sigopt.com/) from 7.4.0 to 7.5.0.

---
updated-dependencies:
- dependency-name: sigopt
  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>
2021-08-23 15:18:59 -07:00
dependabot[bot]
f97a292867
[tune](deps): Bump dask[complete] in /python/requirements/tune (#17997)
Bumps [dask[complete]](https://github.com/dask/dask) from 2021.06.1 to 2021.8.1.
- [Release notes](https://github.com/dask/dask/releases)
- [Changelog](https://github.com/dask/dask/blob/main/docs/release-procedure.md)
- [Commits](https://github.com/dask/dask/compare/2021.06.1...2021.08.1)

---
updated-dependencies:
- dependency-name: dask[complete]
  dependency-type: direct:production
...

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

Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2021-08-23 15:18:37 -07:00
Yi Cheng
5849f80e41
[Core] Fix typo of actor repr (#18011) 2021-08-23 14:33:51 -07:00
Edward Oakes
1a61082ed4
[serve] Remove deprecated endpoints code (#17989) 2021-08-23 13:53:09 -07:00
architkulkarni
f9de05d44c
[Doc] fix incorrect conda example in runtime env doc (#17972) 2021-08-23 13:44:04 -07:00
Amog Kamsetty
4c384df526
fix wheel links (#17973) 2021-08-23 13:43:34 -07:00
Chen Shen
0f894e9cbd
revert ebs cold start (#18010) 2021-08-23 13:40:31 -07:00
chenk008
b9978dd02b
[Core] revert: revert Unified worker starter (#18008) 2021-08-23 13:34:32 -07:00
Yi Cheng
fd71bde9b4
[client] Allow multiple client connections from one driver (#17942) 2021-08-23 13:01:58 -07:00
Sven Mika
d47ed62083
[Testing RLlib] Increase timeout for nightly multi-GPU learning tests. (#17987) 2021-08-23 10:37:50 +02:00
Kai Fricke
fca8af88d2
[release] Fix e2e environment variable passing from pipeline (#18000) 2021-08-23 09:26:37 +02:00
Chen Shen
e369ecab43
Fix EBS cold start in Mac (#18001) 2021-08-22 20:03:59 -07:00
Dmitri Gekhtman
13d5d0f9ef
[autoscaler][hotfix] Update node list after terminating unhealthy nodes (#17992)
* Update nodes; update test.

* consistency

* lint
2021-08-22 18:22:10 -04:00
Clark Zinzow
5ca28b1cc8
[Core] Update Bazel (to 3.4.1), gRPC, boringssl, and absl as a precursor to gRPC streaming PR. (#17903)
* Update Bazel (to 3.4.1), gRPC, boringssl, absl.

* Always reinstall Bazel if needing to upgrade to a new Bazel version.

* Add patch for properly detecting Windows Python headers when building gRPC.

* Add minimum Bazel version check.

* Update docs with new Bazel version.
2021-08-21 11:33:11 -07:00
Sven Mika
494ddd98c1
[RLlib] Replace "seq_lens" w/ SampleBatch.SEQ_LENS. (#17928) 2021-08-21 17:05:48 +02:00
Edward Oakes
b969aa3c80
[dashboard] Don't start dashboard agent when missing dependencies (#17966) 2021-08-21 01:04:21 -07:00
77loopin
c6b24fcb5d
[RayClient] Add the guide for k8s Ingress (#17736)
Co-authored-by: Dmitri Gekhtman <62982571+DmitriGekhtman@users.noreply.github.com>
Co-authored-by: seungjaebaek <seungjaebaek@linecorp.com>
2021-08-20 18:31:03 -07:00
Eric Liang
58e35a21b4
Add split_at_indices() (#17990) 2021-08-20 15:35:22 -07:00
Lixin Wei
05502da271
Add dispatch proxy to event tracker (#17983) 2021-08-20 15:32:10 -07:00
Chen Shen
dac1ba632e
[usability][rfc] ray status show demand summary by default (#17892) 2021-08-20 15:29:37 -07:00
Chris K. W
e3fb9650b2
[Client] Skip client object ref, actor handle, and actor ref dealloc/del if client package has already been cleaned up (#17969) 2021-08-20 15:18:43 -07:00
Chen Shen
31482563c2
[Test] fix-mac-test by avoiding cold start (#17988) 2021-08-20 15:04:29 -07:00
Edward Oakes
3ea5c0dc6b
[serve] Remove deprecated routing code (ServeStarletteRouter) (#17986) 2021-08-20 16:56:45 -05:00
Simon Mo
8236b7412e
[Serve] Mark serve.start beta API (instead of stable) (#17956) 2021-08-20 16:36:48 -05:00
SangBin Cho
cd42d30d7b
[Core] Removing GCS object directory from raylet (#17962) 2021-08-20 12:57:16 -07:00
Chen Shen
3dbb2e0020
change the way test run (#17930) 2021-08-20 11:26:16 -07:00
Chen Shen
89f988e9cc
add dataset shuffle data loader (#17917) 2021-08-20 11:26:01 -07:00
Edward Oakes
30541025e5
[serve] Remove deprecated APIs from code & docs (#17754) 2021-08-20 11:59:45 -05:00
Stephanie Wang
b8fe776638
[core] Fix inlined nested ids (#17834)
* test

* Use ObjectRef instead of ObjectID in nested refs

* java

* doc

* java

* build

* build

* x

* lint

* simplify

* fix
2021-08-20 08:58:29 -07:00
Amog Kamsetty
9416fce91b
[SGD] v2 Tune integration + iterator API (#17839)
* [SGD] implement SGD Trainer.to_tune_trainable

* address some comments

* add RESULT_DUPLICATE

* extract trainable creation logic out of Trainer

* add 1 CPU for driver

* use class attribute to fix serialization issues

* add examples

* add test for tune error

* tune

* test tune_linear

* run_iterator

* add to build file

* Update python/ray/util/sgd/v2/trainer.py

Co-authored-by: matthewdeng <matthew.j.deng@gmail.com>

* Update python/ray/util/sgd/v2/trainer.py

Co-authored-by: matthewdeng <matthew.j.deng@gmail.com>

* address comments

* fix tests & address comments

* resolve merge

* lint

* fix

* add team tag to tests

* fix tests

* lint

Co-authored-by: Matthew Deng <matthew.j.deng@gmail.com>
2021-08-20 08:31:21 -07:00
simonsays1980
60aee4a330
[RLlib] Add example script for bare metal Policy with custom view_requirements. (#17896) 2021-08-20 12:17:13 +02:00
Jingyu-Peng
40330ca439
Fix loading dynamic functions/classes when using code_search_path (#17605) 2021-08-20 17:24:11 +08:00
Antoni Baum
0a1228ef6e
Add configurable autosuspend for connect tests (#17958) 2021-08-20 10:57:41 +02:00
Sven Mika
8248ba531b
[RLlib] Redo #17410: Example script: Remote worker envs with inference done on main node. (#17960) 2021-08-20 08:02:18 +02:00
Eric Liang
236b772465
Revert "[GCS] GCS Based Actor Scheduler (#16580)" (#17941)
This reverts commit a9b4545502.
2021-08-19 21:46:52 -07:00
Eric Liang
661ac4e37b
Remove last traces of ref-counting flag (#17932) 2021-08-19 21:08:13 -07:00
architkulkarni
36c26578a7
[runtime env] [test] Add nightly test to verify Ray wheel URLs are valid (#17938) 2021-08-19 15:48:37 -07:00
Chen Shen
a16a25852a
[Core] fix event race condition (#17947) 2021-08-19 14:20:34 -07:00
matthewdeng
d081ee9d87
[SGD v2] Save checkpoints to disk (#17807)
* [SGD] save checkpoints to disk

* fix test; add logs

* rename log_dir to logdir for consistency with tune

* address comments: add run level directories, add CheckpointConfig

* check for empty strings

Co-authored-by: Antoni Baum <antoni.baum@protonmail.com>

* address comments - refactor CheckpointStrategy, remove run_dir and checkpoint_dir configurability

* fix Trainer docs

* Update python/ray/util/sgd/v2/checkpoint.py

Co-authored-by: Richard Liaw <rliaw@berkeley.edu>

* remove construct_path_with_default

Co-authored-by: Antoni Baum <antoni.baum@protonmail.com>
Co-authored-by: Richard Liaw <rliaw@berkeley.edu>
2021-08-19 14:18:51 -07:00
Sven Mika
a2d96c513a
[RLlib] Expand machine for nightly multi-gpu learning tests. (#17955) 2021-08-19 22:27:30 +02:00
Eric Liang
238941f857
Ray workflow comparison examples + add to tests (#17880) 2021-08-19 12:19:08 -07:00