Commit Graph

3147 Commits

Author SHA1 Message Date
Simon Mo 10c2089061 [Hotfix] Pin Pydantic Version (#11622) 2020-11-02 11:37:53 -08:00
Alan Guo 6aba4ab8f9 Disable validation of cluster config on the cluster to allow for cluster configs with new properties. (#11693) 2020-11-02 11:37:31 -08:00
Alex Wu 7fbf938c3f Version bump 1.0.1 2020-10-30 20:23:59 -07:00
Jack Parker-Holder f137808518 [tune] PB2 (#11466)
Co-authored-by: Sumanth Ratna <sumanthratna@gmail.com>
Co-authored-by: Amog Kamsetty <amogkamsetty@yahoo.com>
Co-authored-by: Amog Kamsetty <amogkam@users.noreply.github.com>
Co-authored-by: Richard Liaw <rliaw@berkeley.edu>
2020-10-30 20:00:52 -07:00
Simon Mo 1f1ea85a5c Fix asyncio plasma integration in cluster mode (#11665) 2020-10-30 20:00:01 -07:00
Raoul Khouri 44a379ee9b [tune] fixed validation for search metrics (#11583)
* fixed validation for search metrics

* formatting

* made error report better

* if only one metric is missing extract it from list

* any can take a generator
2020-10-30 19:59:20 -07:00
Ian Rodney 84fc622659 [yaml] HotFix for correct example full (#11584) 2020-10-23 16:31:45 -07:00
Richard Liaw 395ddb093c [tune] a tiny ptl example (#11497) 2020-10-23 16:14:51 -07:00
Alex Wu c26dbc1612 [Autoscaler] Do not count unmanaged nodes in load metrics (#11458)
* fixedd

* lint

* fixed other test case

* .

Co-authored-by: Alex Wu <alex@anyscale.com>
2020-10-23 16:13:04 -07:00
Kai Yang a2e12ceb2a [Core] Allow creating tasks/actors in a detached actor when driver has exited (#11493)
* Allow creating tasks/actors in a detached actor when driver has exited

* lint

* Address comment
2020-10-23 16:12:26 -07:00
Max Fitton cac4c82c8a [hotfix] Pin node version (fix linux wheel build) (#11532)
Co-authored-by: Max Fitton <max@semprehealth.com>
2020-10-21 19:18:58 -07:00
Edward Oakes dbcb368dea Add --worker-port-list option to ray start (#11481) 2020-10-21 19:18:31 -07:00
Alan Guo 9d765ba740 [autoscaler] Add rsync_exclude and rsync_filter options to cluster config (#11512) 2020-10-21 19:17:34 -07:00
Gekho457 cf1c737895 [autoscaler/AWS] Updated AWS Node Provider threading logic (#11422) 2020-10-21 19:17:00 -07:00
Servon aeea168940 [tune] Update for ZOOpt (#11491)
Co-authored-by: Servon <zewen.li@polixir.ai>
Co-authored-by: Richard Liaw <rliaw@berkeley.edu>
2020-10-20 23:56:20 -07:00
Allen 2fc3237045 [Dashboard] Update dashboard port checking so that we can instantly reuse the dashboard port (#11487)
Co-authored-by: Allen Yin <allenyin@anyscale.io>
2020-10-20 19:19:50 -07:00
Kai Fricke 6d11fb8bc6 [tune] validate function callable in tune.with_parameters() (#11504) 2020-10-20 16:03:24 -07:00
Simon Mo 2c5cb95b42 [Serve] Get ServeHandle on the same node (#11477) 2020-10-20 10:44:23 -07:00
Max Fitton cdca5af53b Revert "[Dashboard] Turn on New Dashboard by Default (#11321)" (#11502)
This reverts commit f500292d41.
2020-10-20 10:53:10 -05:00
DK.Pino 15cbfae8ad fix cross language ut (#11492) 2020-10-20 14:14:40 +08:00
Kai Fricke ed81010b0d [tune] support hierarchical search spaces for hyperopt (#11431)
* support hierarchical search spaces for hyperopt

* Reduce num samples

* Fix prefix
2020-10-19 16:00:08 -07:00
Ian Rodney acbd12eabf [Docker] Set Docker as the Default (#11416) 2020-10-19 10:53:30 -07:00
Max Fitton f500292d41 [Dashboard] Turn on New Dashboard by Default (#11321) 2020-10-19 12:31:11 -05:00
Servon 202b1859ef [tune] Update ZOOpt to better support the latest Ray (#11462)
Co-authored-by: Servon <zewen.li@polixir.ai>
2020-10-19 09:28:11 -07:00
Lee moon soo 8581dd2fb1 [Autoscaler] Staroid node provider followup improvements (#11408) 2020-10-18 00:26:11 -07:00
Ian Rodney 6d6a536601 [Docker] Fix Docker Stop on Teardown (#11400) 2020-10-16 15:45:33 -07:00
DK.Pino 6233cef22c [Placement Group] Remove useless placement group argument in ray remote (#11389)
* remove useless placement_group argument in @ray_remote

* remove
2020-10-16 14:17:18 -07:00
Gekho457 50be2970dc [autoscaler]Type hints for commands.py and sdk.py. (#11354) 2020-10-16 13:45:36 -07:00
mattearllongshot 049985549b Restore uptime timeout to 5 seconds (#11300) 2020-10-15 17:21:11 -07:00
herve-alanaai 436202bcfd [docs] Fix typos in documentation (#11414) 2020-10-15 17:00:48 -07:00
Ian Rodney afd797b896 [docker] Check for GPUs before setting runtime-nvidia (#11418) 2020-10-15 15:43:09 -07:00
Amog Kamsetty 38eb61442b [SGD] Callback API for SGD+Tune (#11316)
Co-authored-by: Richard Liaw <rliaw@berkeley.edu>
2020-10-15 15:22:14 -07:00
Sumanth Ratna 60a4be4a59 [tune] Remove metric and mode kwargs from create_searcher (#11335) 2020-10-14 21:44:36 -07:00
Sumanth Ratna 3fe757391b [tune] Add Basic Variant Generator to search algorithm shim function (#11334)
* Add Basic Variant Generator

* Add 'random' key to SEARCH_ALG_IMPORT

Co-authored-by: Kai Fricke <kai@anyscale.com>

Co-authored-by: Kai Fricke <kai@anyscale.com>
2020-10-14 21:41:47 -07:00
Vishnu Deva 00e0f14c6f [tune] restore trials when sync_on_checkpoint is False (#11355)
Co-authored-by: Richard Liaw <rliaw@berkeley.edu>
2020-10-14 19:09:34 -07:00
Kai Fricke f7120d2a18 [tune] Make metrics parameter optional in pytorch lightning integration (#11402) 2020-10-14 17:50:34 -07:00
SangBin Cho 666fcde8ca [Placement group] Input validation (#11152)
* Add a basic input validation.

* Addressed code review.
2020-10-14 13:56:41 -07:00
Ameer Haj Ali a10e36ca04 Make the logging of gc.collect() freed refs appear in DEBUG not INFO (#11353) 2020-10-14 13:14:35 -07:00
Alex Wu 7466ce82df [Autoscaler] Placement group autoscaling (#11243) 2020-10-14 13:11:46 -07:00
Kai Yang abc6126814 [Java] Release actor instance reference when Ray.exitActor() is invoked (#11324) 2020-10-14 13:12:59 +08:00
Simon Mo 866193b01c Fix cluster yaml for serve benchmarks (#11383)
- Separate out single node and multiple node yamls
- Remove cluster_synced_files, somehow it breaks for me
2020-10-13 19:30:18 -07:00
Max Fitton cd9dcfca0d [Dashboard] CPU/GPU usage details in actor pane (#11269) 2020-10-13 20:23:23 -05:00
Amog Kamsetty 933cf6675c [Tune] Changes for Pytorch Lightning 1.0 (#11375) 2020-10-13 15:50:11 -07:00
J Seppänen 63fa0a53a3 [k8s] Fix kubernetes cloud cluster example configuration (#11364) 2020-10-13 12:28:55 -07:00
SangBin Cho 56f69543d0 Try to deflake test_failure (#11293) 2020-10-12 12:03:36 -05:00
Ameer Haj Ali 06fe690682 [autoscaler] Limit max launch concurrency per node type (#11242) 2020-10-12 09:45:52 -07:00
Alex Wu 175fc41fbc [Autoscaler] Account for resource backlog size (#11261) 2020-10-12 09:43:48 -07:00
Simon Mo 0d09a17c64 Skip set_result if the future is done (#11256) 2020-10-11 22:33:58 -07:00
Richard Liaw 56f858ed1a [tune][docs/util] gputil check, docs (#11260)
Co-authored-by: Amog Kamsetty <amogkam@users.noreply.github.com>
2020-10-10 00:54:31 -07:00
Gekho457 48db6f8858 [autoscaler/k8s] namespace permissions problem (#11270) 2020-10-09 19:22:20 -05:00