Commit Graph

220 Commits

Author SHA1 Message Date
Kai Fricke 9f5986ee58 [tune] logger migration to ExperimentLogger classes (#11984) 2020-11-16 15:08:37 -08:00
Kai Fricke 8609e2dd90 [tune] refactor verbosity levels (#11767)
Co-authored-by: Richard Liaw <rliaw@berkeley.edu>
2020-11-16 10:32:53 -08:00
Keqiu Hu a50128079d [tune/placement group] dist. training placement group support (#11934)
Co-authored-by: Richard Liaw <rliaw@berkeley.edu>
2020-11-16 01:11:39 -08:00
Richard Liaw 8b3f79f307 [tune] refactor and add examples (#11931) 2020-11-14 20:43:28 -08:00
Kai Fricke 02c02369ca [tune] Fix hpo randint limits (#11946)
Co-authored-by: Sumanth Ratna <sumanthratna@gmail.com>
2020-11-12 08:45:49 -08:00
Keqiu Hu 0c1bdaef59 [tune] TensorFlow Distributed Trainable (#11876)
Co-authored-by: Richard Liaw <rliaw@berkeley.edu>
2020-11-10 14:59:08 -08:00
Eric Liang 9b8218aabd [docs] Move all /latest links to /master (#11897)
* use master link

* remae

* revert non-ray

* more

* mre
2020-11-10 10:53:28 -08:00
Kai Fricke 287aba6dc3 [tune] schedulers: Add test for context finalization (#11889) 2020-11-09 11:37:05 -08:00
Kai Fricke 88be1ea20b [tune] Handle infinite and NaN values (#11835) 2020-11-09 11:18:31 -08:00
Kai Fricke 603accf1c2 [tune] logger refactor part 3: Add ExperimentLogger class (#11749) 2020-11-05 08:55:38 -08:00
Kai Fricke 007634fd1b [tune] logger refactor part 2: Add SyncerCallback (#11748)
Co-authored-by: Richard Liaw <rliaw@berkeley.edu>
2020-11-03 21:04:40 -08:00
Kai Fricke f7b19c41e3 [tune] logger refactor part 1: move classes and utilities to own files (#11746)
* [tune] logger refactor part 1: move classes and utilities to own files

* Fix circular dependency

* Remove uneeded pretty print copy

* Apply suggestions from code review
2020-11-03 07:48:09 -08:00
Kai Fricke ba63ded311 [tune] better error when metric or mode unset in search algorithms (#11646) 2020-10-28 13:17:59 -07:00
Richard Liaw 58891551d3 [tune] make tests faster + fix flaky test (#10264) 2020-10-28 13:14:54 -07:00
Kai Fricke 1a1ff28d18 [tune] allow tune search spaces to be passed to search algorithms (#11503) 2020-10-26 12:33:13 -07:00
Richard Liaw 1b357533b1 [tune] Try to enable PTL, SKlearn tests (#11542) 2020-10-24 01:08:46 -07:00
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
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
Kai Fricke f7120d2a18 [tune] Make metrics parameter optional in pytorch lightning integration (#11402) 2020-10-14 17:50:34 -07:00
Amog Kamsetty 933cf6675c [Tune] Changes for Pytorch Lightning 1.0 (#11375) 2020-10-13 15:50:11 -07:00
Kai Fricke b450cb030a [tune] reuse actors for function API (#11230)
Co-authored-by: Kristian Hartikainen <kristian.hartikainen@gmail.com>
2020-10-08 16:15:02 -07:00
Richard Liaw 74e9647ec3 [tune] deactivate flaky test for now (#11284) 2020-10-08 15:40:04 -07:00
Amog Kamsetty 1027bfd4b8 [Tune, Ray SGD] Update PTL integrations (#11271) 2020-10-08 13:43:07 -07:00
Kai Fricke bdf647c4ec [tune] docker syncer (#11035)
* Add DockerSyncer

* Add docs

* Update python/ray/tune/integration/docker.py

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

* Updated docs

* fix dir

* Added docker integration test

* added docker integration test to bazel build

* Use sdk.rsync API

Co-authored-by: Richard Liaw <rliaw@berkeley.edu>
2020-10-01 11:59:23 -07:00
Kai Fricke c77cfaa5ad [tune] use dated experiment dir per default (#11104) 2020-09-30 14:43:59 -07:00
Kai Fricke 10015e60fb [tune] fix flaky callback test (#11102) 2020-09-29 09:44:32 -07:00
Kai Fricke e7315b0856 [tune] Callbacks for tune runs (#11001) 2020-09-27 16:50:07 -07:00
Eric Liang 609c1b8acd Start moving ray internal files to _private module (#10994) 2020-09-24 22:46:35 -07:00
Amog Kamsetty ee85cb31a5 [Tune] Fix Memory Leak (#10989)
Co-authored-by: Richard Liaw <rliaw@berkeley.edu>
2020-09-24 20:26:55 -07:00
Kai Fricke d9c4dea7cf [tune] strict metric checking (#10972) 2020-09-24 10:00:48 -07:00
Kai Fricke 5921e87ecd [tune] Only add new trial when there is no pending trial (#10979) 2020-09-23 11:08:12 -07:00
Kai Fricke 6247740b94 [tune] sort running trials to top in status table (#10926)
Co-authored-by: Richard Liaw <rliaw@berkeley.edu>
2020-09-21 13:39:51 -07:00
Eric Liang 6a227ae501 [autoscaler] Split autoscaler interface public private (#10898) 2020-09-18 18:16:23 -07:00
Kai Fricke 508cfa3540 [tune] Support yield and return statements (#10857)
* Support `yield` and `return` statements in Tune trainable functions

* Support anonymous metric with ``tune.report(value)``

* Raise on invalid return/yield value

* Fix end to end reporter test
2020-09-17 20:18:35 -07:00
Kai Fricke ee99c919e3 [tune] lazy trials (#10802)
* Lazily fill trial queue

* Update interface

* Update end to end reporter test

* Removed `next_trials()` method

* Lint

* Print total number of samples to be generated in progress reporter. Allow infinite samples.

* Nit check
2020-09-17 08:51:46 -07:00
Richard Liaw d3feb83053 [tune] check for running session (#10840) 2020-09-16 18:55:11 -07:00
Kai Fricke 0865d68466 [tune] convert fallback representation to numbers in wandb integration (#10799) 2020-09-15 11:47:11 -07:00
Ameer Haj Ali 6edacb22b8 Fix abstraction violations in command_runner interface (#10715)
* Fix abstraction violations in command_runner interface

* user guide

* lint

* breaking abstraction in commands

* extra initialization commands

* more cleanup

* small fixes

* fix test_integration_kubernetes.py

* lint

Co-authored-by: root <root@ip-172-31-28-155.us-west-2.compute.internal>
Co-authored-by: Ameer Haj Ali <ameerhajali@Ameers-MacBook-Pro.local>
2020-09-14 20:28:38 -07:00
Ian Rodney 5bc2ba38fd [docker] Detect CPUs in container correctly (#10507)
Co-authored-by: simon-mo <simon.mo@hey.com>
Co-authored-by: Richard Liaw <rliaw@berkeley.edu>
Co-authored-by: Alex Wu <itswu.alex@gmail.com>
2020-09-13 23:40:48 -07:00
Richard Liaw 6662efd74b [tune/rllib] revert removal of queue-trials (#10744) 2020-09-11 14:13:20 -07:00
Kai Fricke 7eaf063f29 [tune] wrapper function to pass arbitrary objects through the object store to trainables (#10679) 2020-09-10 17:39:44 -07:00
Richard Liaw 153813936b [tune] auto infer metrics (#10663)
Co-authored-by: Kai Fricke <krfricke@users.noreply.github.com>
Co-authored-by: Kai Fricke <kai@anyscale.com>
2020-09-09 09:53:47 -07:00
Kai Fricke d7c7aba99c [tune] Tune experiment analysis improvements (#10645)
Co-authored-by: Richard Liaw <rliaw@berkeley.edu>
2020-09-08 21:00:52 -07:00
Kai Fricke 756a9ea641 [tune] add mode/metric parameters to tune.run (#10627)
Co-authored-by: Richard Liaw <rliaw@berkeley.edu>
2020-09-08 17:06:21 -07:00
Kai Fricke 87c4f36f02 [tune] Added timeout parameter to tune.run(), (#10642) 2020-09-08 15:38:28 -07:00
Richard Liaw 5851e893ee [tune] More robust resolution/detection of signature (#10365)
Co-authored-by: Amog Kamsetty <amogkam@users.noreply.github.com>
2020-09-08 11:38:16 -07:00
Kai Fricke 088f8ebb69 [tune] Add algorithms for search space conversion (#10621) 2020-09-07 13:44:16 -07:00
Kai Fricke c5e9bafe15 [tune] Fix flaky test in test_sample (#10602) 2020-09-06 10:29:48 -07:00
Richard Liaw 551c597312 [tune] API revamp fix (#10518) 2020-09-05 15:34:53 -07:00
Sumanth Ratna 54215ff287 [tune] implement shim instantiation (#10456)
* Create ray.tune.suggest.create.create_scheduler

* Update __init__.py

* Resolve conflict in __init__.py

* Create ray.tune.schedulers.create.create_scheduler

* Update __init__.py

* Move create_scheduler to tune.schedulers.__init__

* Move create_searcher to tune.suggest.__init__

* Delete tune.suggest.create

* Delete tune.schedulers.create

* Update imports for shim functions in tune.__init__

* Remove shim from tune.suggest.__init__.__all__

* Remove shim from tune.schedulers.__init__.__all__

* Add ShimCreationTest

* Move ShimCreationTest to test_api

* Delete test_shim.py

* Add docstring for ray.tune.create_scheduler

* Add docstring to ray.tune.create_searcher

* Fix typo in ray.tune.create_scheduler docstring

* Fix lint errors in tune.schedulers.__init__

* Fix lint errors in tune.suggest.__init__

* Fix lint errors in tune.suggest.__init__

* Fix lint errors in tune.schedulers.__init__

* Fix imports in test_api

* Fix lint errors in test_api

* Fix kwargs in create_searcher

* Fix kwargs in create_scheduler

* Merge branch 'master' into shim-instantiation

* Update use-case in docs in tune.create_scheduler

* Update use-case in docs in tune.create_searcher

* Remove duplicate pytest run from test_api

* Add check to create_searcher


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

* Add check to create_scheduler

* lint

* Compare types of instances in test_api

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

* Add tune.create_searcher to docs

* Fix doc build

* Fix tests

* Add tune.create_scheduler to docs

* Fix tests

* Fix lint errors

* Update Ax search for master

* Fix metric kwarg for Ax in test_api

* Fix doc build

* Fix HyperOptSearch import in test_api

* Fix HyperOptSearch import in create_searcher

Co-authored-by: Richard Liaw <rliaw@berkeley.edu>
2020-09-05 09:36:42 -07:00