Kai Fricke
9b0d804eed
[tune] Add documentation for reproducible runs (setting seeds) ( #18849 )
...
Co-authored-by: Antoni Baum <antoni.baum@protonmail.com>
2021-09-24 10:57:31 +01:00
Chen Shen
7c99aae033
[dataset][nightly-test] add pipelined ingestion/training nightly test
2021-09-23 20:39:03 -07:00
Simon Mo
565131a854
[Serve] Support http_location=FixedNumber ( #18731 )
2021-09-23 15:59:12 -07:00
Simon Mo
5aa1e08633
[Serve] Exit run_forever when actor shutdown ( #18820 )
2021-09-23 15:17:31 -07:00
Yi Cheng
b5ccee6ad3
Skip failed actor test ( #18815 )
2021-09-23 11:02:02 -07:00
Kai Fricke
2d46e0e14b
[tune] Fix Analysis.dataframe() documentation and enable passing of mode=None ( #18850 )
2021-09-23 18:27:54 +01:00
Jiajun Yao
cc84f18176
Increase disk for long running distributed tests ( #18855 )
2021-09-23 17:52:35 +01:00
Stephanie Wang
7b1e594412
[core] Fix bug in ref counting protocol for nested objects ( #18821 )
...
* Fix assertion crash
* test, lint
* todo
* tests
* protocol
* test
* fix
* lint
* header
* recursive
* note
* forward test
* lock
* lint
* unneeded check
2021-09-23 09:45:12 -07:00
Alex Wu
5d57eed598
[Workflow] Serialization cleanup ( #18328 )
...
* notes
* notes
* .
* seems to work?
* .
* seems to work
* needs tests
* needs tests
* parallelize uploads
* fixed
* fixed
* .
* dumb test
* .
* .
* fix festsg
* .
* works
* .:
* .
* .
* .
* Update common.py
* .
* almost removed special case for inputs
* lint
* lint
* .
* handle edge case
* .
* .
* lint
* needs dedupe
* needs dedupe
* still need to not leak cache
* still need to not leak cache
* probably fails edge cases?
* probably fails edge cases?
* works?
* cleanup
* passes test?
* ???
* done?
* may work?
* may work?
* .
* .
* Revert "."
This reverts commit 6aee40630637783d1756e226861b518668112337.
* Revert "."
This reverts commit 040a0e59e731d1f4e3b85ca2153474fc97963ae8.
* Revert "may work?"
This reverts commit fc26b54627c3c72dfdbaf0e79ba89d7503db4a94.
* Revert "may work?"
This reverts commit 85f48bb11a5c1764ef2cf3701ec41eb948fc7fc1.
* Revert "done?"
This reverts commit 573f4e0cb98417494b30c7a36987391d9bb8d064.
* passs tests
* lint
* cleanup
* bug fix
* bug fix
* print
Co-authored-by: Alex Wu <alex@anyscale.com>
2021-09-23 09:18:59 -07:00
Carl Assmann
882f7d3863
[tune] OptunaSearch: check compatibility of search space with evaluated_rewards ( #18625 )
...
Co-authored-by: Antoni Baum <antoni.baum@protonmail.com>
Co-authored-by: Kai Fricke <krfricke@users.noreply.github.com>
2021-09-23 16:00:11 +01:00
Amog Kamsetty
99b1d8c95f
[SGD] Update Docs ( #18839 )
2021-09-23 07:52:57 -07:00
Guyang Song
739cf64115
[C++ API] support head_args config in C++ API ( #18709 )
2021-09-23 19:30:53 +08:00
Sven Mika
61a1274619
[RLlib] No Preprocessors (part 2). ( #18468 )
2021-09-23 12:56:45 +02:00
Sven Mika
a2a077b874
[RLlib] Faster remote worker space inference (don't infer if not required). ( #18805 )
2021-09-23 10:54:37 +02:00
Antoni Baum
361cae4d1c
[tune] Add save and restore methods for searchers that were missing it & test ( #18760 )
2021-09-23 09:45:47 +01:00
Eric Liang
2c15215833
Implement zip() function for dataset ( #18833 )
2021-09-23 00:12:29 -07:00
Sven Mika
a96dbd885b
[RLlib] Reinstate trajectory view API tests. ( #18809 )
2021-09-23 08:31:51 +02:00
Guyang Song
237a2ade76
[wheel][cpp] recover cpp extra ( #18597 )
2021-09-23 12:10:03 +08:00
Amog Kamsetty
d354161528
[SGD] Link ray.sgd namespace to ray.util.sgd.v2 ( #18732 )
...
* wip
* add symlink
* update
* remove from init
* no require tune
* try fix
* change
* * import
* fix docs
* address comment
2021-09-22 18:49:41 -07:00
mwtian
e41109a5e7
[Client] Use async rpc for remote call and actor creation ( #18298 )
...
* Use async rpc for remote calls, task and actor creations.
* fix
* check placement
* check placement group. wait for id in destructor
* fix
* fix exception in destructor
* Add test
* revert change
* Fix comment
* fix
2021-09-22 18:30:50 -07:00
Yi Cheng
8dd3057644
Revert "[test] add unit test for PR #17634 ( #18585 )" ( #18830 )
...
This reverts commit 73c3cff18b
2021-09-22 16:51:02 -07:00
Amog Kamsetty
00dd190df9
[SGD] Retry sgd.local_rank() ( #18824 )
...
* finish
* fix
* wip
* address comment
* update
* fix test
* fix failing test
* address comments
* fix test
* fix
2021-09-22 15:48:38 -07:00
Yi Cheng
73c3cff18b
[test] add unit test for PR #17634 ( #18585 )
2021-09-22 14:39:30 -07:00
gjoliver
e6511bcf56
Revert "Upgrade default bazel installation to ver 4.2.1 ( #18714 )" ( #18825 )
2021-09-22 13:54:48 -07:00
Amog Kamsetty
d9b166252b
Revert "[SGD] sgd.local_rank" ( #18822 )
2021-09-22 13:50:00 -07:00
Amog Kamsetty
42c925ca0a
[Docs] Fix ray[default] Wheel install instruction ( #18819 )
2021-09-22 12:53:08 -07:00
Sven Mika
93208bb087
[RLlib] Increase size of (very flakey) action_masking example script test. ( #18816 )
2021-09-22 21:48:01 +02:00
Clark Zinzow
a3f40236d0
[Repo Config] Allow blank issues. ( #18800 )
2021-09-22 11:39:59 -07:00
Chen Shen
9b1cd5d1ad
Disable spill test on macOS ( #18801 )
2021-09-22 09:57:53 -07:00
Amog Kamsetty
39bcbe03bc
[SGD] sgd.local_rank ( #18686 )
...
* finish
* fix
* wip
* address comment
* update
* fix test
* fix failing test
* address comments
* fix test
2021-09-22 08:10:49 -07:00
Kai Fricke
bbb207c36e
[sgd/v1] Add API annotations ( #18790 )
...
* [sgd/v1] Add API annotations
* Remove unnecessary annotations
2021-09-22 08:10:28 -07:00
Sven Mika
5611150b1a
Increase rllib stress tests timeout for smoke test ( #18810 )
2021-09-22 14:30:42 +01:00
Qing Wang
3ad1553b34
[Java] Remove API setJvmOptions(String). ( #18664 )
2021-09-22 20:00:49 +08:00
Kai Fricke
2cbf326410
[ci/release] store buildkite artifacts on buildkite ( #18712 )
2021-09-22 11:35:59 +01:00
Kai Fricke
f86fc277d6
[tune/rllib] Only disable ipython in remote actors ( #18789 )
2021-09-22 11:05:06 +01:00
gjoliver
eb3620898c
Upgrade default bazel installation to ver 4.2.1 ( #18714 )
2021-09-22 00:24:41 -07:00
Eric Liang
cf0bd00cc2
Improve the error message for failed task/actor imports on workers ( #18792 )
2021-09-21 19:49:59 -07:00
Sven Mika
698b4eeed3
[RLlib] POC: Separate losses for APPO/IMPALA. Enable TFPolicy to handle multiple optimizers/losses (like TorchPolicy). ( #18669 )
2021-09-21 22:00:14 +02:00
Antoni Baum
3106fc5365
[tune] Depreciate max_concurrent in TuneBOHB ( #18770 )
...
Co-authored-by: Kai Fricke <krfricke@users.noreply.github.com>
2021-09-21 19:17:19 +01:00
architkulkarni
aa6625e62a
[Serve] gate __del__ call behind hasattr check ( #18773 )
2021-09-21 10:48:40 -07:00
Antoni Baum
f4666f3a6d
[tune] Add on_trial_result to ConcurrencyLimiter ( #18766 )
2021-09-21 15:30:02 +01:00
Antoni Baum
ca3fabc4cb
[tune] Ensure arguments passed to tune remote_run match ( #18733 )
2021-09-21 15:29:29 +01:00
Yi Cheng
fc6a739e4b
[nightly] Deflaky nightly test many_nodes_actor_test ( #18582 )
2021-09-20 22:43:48 -07:00
Clark Zinzow
0704b825ff
[Datasets] Add spread resource prefix for manual round-robin resource-based task load balancing. ( #18776 )
2021-09-20 22:41:11 -07:00
Eric Liang
361a13602c
Actor repr for log prefix should be computed after init, not before ( #18749 )
2021-09-20 21:34:53 -07:00
DK.Pino
d329101469
Revert Revert "[Placement Group] Support infeasible placement groups for Placement Group." ( #18735 )
...
* fix conflict
* cxx lint
2021-09-20 20:18:12 -07:00
Yi Cheng
07babd807c
Revert "Revert "[core] Async submitting actor registerring ( #18009 )" ( #18719 )" ( #18722 )
2021-09-20 19:17:00 -07:00
Ameer Haj Ali
9efbd80733
[core] avoid scheduling on gpu nodes by default ( #18743 )
...
* [core] avoid scheduling on gpu nodes by default
* Fix cluster_task_manager_test tests.
Made most tests in cluster_task_manager_test not use GPU on the head
node.
Also added another test to scheduling_policy_test.
Co-authored-by: Sasha Sobol <sasha@asobol.com>
2021-09-20 17:38:40 -07:00
Sasha Sobol
65c1c8bb9e
Add an integration test for scheduler_avoid_gpu_nodes ( #18763 )
2021-09-20 17:20:42 -07:00
Jiao
9bb4a87031
[runtime_env] Add experimental job yaml ( #18768 )
2021-09-20 18:00:25 -05:00
