Commit graph

5236 commits

Author SHA1 Message Date
Guyang Song
3bc5f0501f
fix WaitPlacementGroupReady API (#18464) 2021-09-13 14:07:40 +08:00
Jiajun Yao
ae10a80d5e
Fix async actor worker process leak after calling ray.actor.exit_actor() (#18526) 2021-09-12 11:09:12 -07:00
Yi Cheng
15d67aa775
Support workflows step names via decorator (#18520) 2021-09-11 13:39:07 -07:00
Qing Wang
371f03fa48
Remove dynamic resource from client side. (#18514) 2021-09-11 10:39:59 -07:00
Chong-Li
d314d0c10e
[GCS] Fix the Windows build of GCS actor scheduling (#18012) 2021-09-10 17:17:25 -07:00
Alex Wu
1587eb22f0
[workflow] Dedupe object reference uploads (#18438)
* maybe works

* ?

* .

* seems to work

* seems to work

* .

* .

* lint

* address comments

* .

* test

* test

* .

* works?

* cleanup

* cleanup

* cleanup

* cleanup

* fix test + cleanup

* lint

Co-authored-by: Alex Wu <alex@anyscale.com>
2021-09-10 16:08:11 -07:00
dependabot[bot]
30012c990f
[tune](deps): Bump matplotlib in /python/requirements/tune (#18025)
Bumps [matplotlib](https://github.com/matplotlib/matplotlib) from 3.4.2 to 3.4.3.
- [Release notes](https://github.com/matplotlib/matplotlib/releases)
- [Commits](https://github.com/matplotlib/matplotlib/compare/v3.4.2...v3.4.3)

---
updated-dependencies:
- dependency-name: matplotlib
  dependency-type: direct:production
  update-type: version-update:semver-patch
...

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

Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2021-09-10 16:00:16 -07:00
dependabot[bot]
42794c7a3e
[tune](deps): Bump pytorch-lightning in /python/requirements/tune (#18359)
Bumps [pytorch-lightning](https://github.com/PyTorchLightning/pytorch-lightning) from 1.4.3 to 1.4.5.
- [Release notes](https://github.com/PyTorchLightning/pytorch-lightning/releases)
- [Changelog](https://github.com/PyTorchLightning/pytorch-lightning/blob/master/CHANGELOG.md)
- [Commits](https://github.com/PyTorchLightning/pytorch-lightning/compare/1.4.3...1.4.5)

---
updated-dependencies:
- dependency-name: pytorch-lightning
  dependency-type: direct:production
  update-type: version-update:semver-patch
...

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

Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2021-09-10 15:59:26 -07:00
Lixin Wei
88de723e62
Fix flaky test_gcs_fault_tolerance.py (#18493) 2021-09-10 11:34:07 -07:00
Chris K. W
6f94d0f3c9
[client] Use application specific error code to propagate ray errors (#18278)
* Raise decoded exception if generated by grpc lib

* Switch to missing client_id error to FAILED_PRECONDITION

* switch to ABORTED

* fix comment

* fix decode_exception comment
2021-09-10 09:49:03 -07:00
Ameer Haj Ali
ead02b21b9
[client] Fix exception error message (#18485) 2021-09-10 14:34:31 +03:00
Guyang Song
03a2c69a8a
Don't add ray-cpp wheel to extras by default (#18251) 2021-09-10 09:56:51 +01:00
xwjiang2010
ae689ecc6b
[tune] Add optional Experiment to Searcher/SearchAlgo. (#17724) 2021-09-10 09:30:18 +01:00
Edward Oakes
2fcfea10b3
[runtime_env] Move URI deletion logic to the agent, remove util worker code (#18471) 2021-09-10 00:13:32 -07:00
Yi Cheng
f2d8f23fb6
[workflow] Define default __getstate__ and __setstate__ (#18459) 2021-09-09 23:04:00 -07:00
Yi Cheng
965c55fe1b
[workflow] set max retry to 3 (#18477) 2021-09-09 23:03:24 -07:00
SangBin Cho
7b2ed4c1f8
[Placement group] Placement group scheduling hangs due to creation/removal race condition (#18419) 2021-09-09 20:39:01 -07:00
Amog Kamsetty
d3d8120db3
[SGD] Fix shutdown hang on macOS Python 3.7 (#18473) 2021-09-09 15:32:52 -07:00
Eric Liang
4d2065352b
Increase dataset read parallelism by default (#18420) 2021-09-09 15:07:49 -07:00
Kai Fricke
395976c8a1
[tune] Never block for results (#18391)
* [tune] Never block for results

* Fix tests

* Block in tests

* Add comment to test
2021-09-09 12:08:00 -07:00
architkulkarni
0126837868
[ray client] [runtime env] Print error logs in driver upon connection failure (#18451) 2021-09-09 13:50:55 -05:00
Simon Mo
d477fd7205
[Serve] Make @ingress accept any ASGI app (#15464) 2021-09-09 13:49:37 -05:00
Simon Mo
51bb7c6da8
[Serve] Allow method redefinition for FastAPI (#18453) 2021-09-09 13:48:58 -05:00
Edward Oakes
791abd4f04
[serve] Remove root_url from start docstring (#18472) 2021-09-09 13:46:56 -05:00
Zhi Lin
2fcd1bcb4b
[Dataset] implement from_spark, to_spark and some optimizations (#17340) 2021-09-09 11:43:47 -07:00
SangBin Cho
fdd52106bf
[Placement group] Do not report ready task demand (#18463) 2021-09-09 11:42:12 -07:00
Nikita Vemuri
0f562874b9
[serve] Use environment variable for root_url from runtime env (#18269) 2021-09-09 12:12:35 -05:00
architkulkarni
5affb074aa
[Test] deflake test_runtime_env.py::test_no_spurious_worker_startup (#17809) 2021-09-08 16:35:08 -07:00
Clark Zinzow
c0ea2755a0
Fix iter_batches dropping batches when prefetching. (#18441) 2021-09-08 15:37:38 -07:00
Clark Zinzow
6fc91fd47e
Create directory on write if it doesn't exist. (#18435) 2021-09-08 15:31:06 -07:00
Edward Oakes
f0555f88d6
[runtime_env] Move worker process startup logic to context (#18341) 2021-09-08 17:08:27 -05:00
Antoni Baum
dd6abed6ce
[tune] Fix an edge case where DurableTrainable would not delete checkpoints in remote storage (#18318)
Co-authored-by: Kai Fricke <krfricke@users.noreply.github.com>
2021-09-08 15:00:09 -07:00
Ian Rodney
c91e0eb065
[Dashboard] Increase Actor Snapshot Size (#18433) 2021-09-08 12:06:33 -07:00
Sasha Sobol
f76f14fedf
[client] pass _credentials down from init (#18425) 2021-09-08 10:30:26 -07:00
Clark Zinzow
b30c41759d
[Datasets] Adds tensor column support (tensors-in-tables) via Pandas/Arrow extension types/arrays. (#18301) 2021-09-08 10:09:01 -07:00
mwtian
e427e4a467
Fix flakiness in test_proxy_manager_internal_kv (#18416) 2021-09-08 15:46:45 +03:00
Kai Fricke
dac3a8bc8e
[setup] Upstream conda patches (#17575)
Co-authored-by: Vasilij Litvinov <vasilij.n.litvinov@intel.com>
2021-09-08 10:37:17 +01:00
Edward Oakes
56adaa32f1
[serve] Better logging for exceptions in backend_state.update() (#18402) 2021-09-07 21:40:41 -05:00
Simon Mo
a29da81cfc
Revert "Revert "Fix tracing bug when actors are defined before connecting to …" (#16122) 2021-09-07 16:19:49 -07:00
Edward Oakes
f2afb08125
[runtime_env] Don't modify passed runtime_env dictionary when validating (#18404) 2021-09-07 16:14:28 -07:00
Lada Kunc
1a72c49009
[serve] Fix get_handle execution from threads (#18198) 2021-09-07 14:49:36 -07:00
Guyang Song
f104a5aad7
[docs] Fix cpp wheel description (#18386) 2021-09-07 15:45:04 -05:00
xwjiang2010
64c2f86a22
[Tune] Respect default_resources during Trial.reset(). (#18209) 2021-09-07 19:14:44 +01:00
Clark Zinzow
26b2720915
Add test coverage for writing to fsspec filesystems. (#18394) 2021-09-07 10:16:59 -07:00
Jiajun Yao
2740d28fad
[client] Increase timeout for ProxyManager.get_channel (#18350) 2021-09-07 11:06:17 -05:00
Sven Mika
cabaa3b3c6
[RLlib Testing] Add A3C/APPO/BC/DDPPO/MARWIL/CQL/ES/ARS/TD3 to weekly learning tests. (#18381) 2021-09-07 11:48:41 +02:00
Jiajun Yao
64040a90a5
Datasets schema should match the columns selection for Parquet (#18361) 2021-09-07 00:41:26 -07:00
Sasha Sobol
f24ccf475e
[client] Add a grpc.ChannelCredentials argument to ray.init (#18365)
Co-authored-by: Thomas Desrosiers <thomas@anyscale.com>
2021-09-07 00:17:13 -07:00
Kai Fricke
f3a3a4bc92
[tune] Queue more than more actor/placement group (#18338) 2021-09-06 09:41:08 -07:00
Eric Liang
cbdafa0b63
[doc] Fix various workflow doc bugs (#18357) 2021-09-06 01:39:08 -07:00