Commit Graph

6775 Commits

Author SHA1 Message Date
Stephanie Wang d95c8b8a41 [core][new scheduler] Move tasks from ready to dispatch to waiting on argument eviction (#13048)
* Add index for tasks to dispatch

* Task dependency manager interface

* Unsubscribe dependencies and tests

* NodeManager

* Revert "Add index for tasks to dispatch"

This reverts commit c6ccb9aa306e00f80d34b991055e4e83872595ea.

* tmp

* Move back to waiting if args not ready

* update
2020-12-23 09:33:43 -08:00
Antoni Baum a4f2dd2138 [Tune]Add integer loguniform support (#12994)
* Add integer quantization and loguniform support

* Fix hyperopt qloguniform not being np.log'd first

* Add tests, __init__

* Try to fix tests, better exceptions

* Tweak docstrings

* Type checks in SearchSpaceTest

* Update docs

* Lint, tests

* Update doc/source/tune/api_docs/search_space.rst

Co-authored-by: Kai Fricke <krfricke@users.noreply.github.com>

Co-authored-by: Kai Fricke <krfricke@users.noreply.github.com>
2020-12-23 09:27:16 -08:00
Ameer Haj Ali d37e2c3a20 [joblib] Fix flaky joblib test. (#13046) 2020-12-23 10:43:34 -06:00
Sven Mika 1e74187179 [RLlib] TorchPolicies: Accessing "infos" dict in train_batch causes TypeError. (#13039) 2020-12-23 11:30:50 -05:00
Sven Mika 670d083a56 [RLlib] Fix broken unity3d_env import in example server script. (#13040) 2020-12-23 11:29:58 -05:00
DK.Pino 6e19facc7f [GCS] Delete redis gcs client and redis_xxx_accessor (#12996) 2020-12-23 20:31:46 +08:00
Barak Michener c4e273920f [ray_client]: Insert decorators into the real ray module to allow for client mode (#13031) 2020-12-22 22:51:45 -08:00
fangfengbin 646c4201ac [GCS]Decouple gcs resource manager and gcs node manager (#13012) 2020-12-23 11:25:01 +08:00
fyrestone 62a5832007 [Dashboard] Add GET /logical/actors API (#12913) 2020-12-23 11:14:23 +08:00
Simon Mo bc68260144 [Serve] Handle Bug Fixes (#12971) 2020-12-22 19:13:16 -08:00
Eric Liang 81d3cbaa77 Add "beta" documentation for enabling object spilling manually (#13047) 2020-12-22 16:08:41 -08:00
Alex Wu a79c9fcac3 [release tests] test_many_tasks fix (#12984) 2020-12-22 11:05:33 -08:00
Sven Mika 01faeabc17 [RLlib] Issue 12789: RLlib throws the warning "The given NumPy array is not writeable" (#12793) 2020-12-22 09:28:07 -05:00
Alex Wu ea8d782be1 [core] Pull Manager exponential backoff (#13024) 2020-12-21 19:17:51 -08:00
Edward Oakes b52cce6632 [serve] Refactor SystemState into EndpointState and BackendState (#13018) 2020-12-21 20:39:13 -06:00
Sven Mika d5604eaba3 [RLlib] Attention nets PyTorch support and cleanup (using traj. view API). (#12029) 2020-12-21 18:38:34 -08:00
Eric Liang 8068041006 Don't release resources during plasma fetch (#13025) 2020-12-21 18:32:40 -08:00
Edward Oakes 015a0f9935 [serve] Rename replica_tag -> replica in metrics for consistency (#13022) 2020-12-21 17:19:39 -06:00
Eric Liang 03a5b90ed6 Revert "Revert "Increase the number of unique bits for actors to avoi… (#12990) 2020-12-21 15:16:42 -08:00
architkulkarni 8b4b4bf0a2 [Serve] Migrate from Flask.Request to Starlette Request (#12852) 2020-12-21 15:34:15 -06:00
Hao Zhang 5b48480e29 [Collective][PR 3/6] Other collectives (#12864) 2020-12-21 12:48:00 -08:00
Barak Michener 43b9c7811e [ray_client] add client microbenchmarks (#13007) 2020-12-21 12:17:44 -08:00
Ameer Haj Ali 5e2b850836 [autoscaler] Fixes max_workers bug. (#13008) 2020-12-21 10:30:03 -08:00
roireshef ef95db51e1 [RLlib] Arbitrary input to value() when not using GAE (#12941) 2020-12-21 12:19:33 -05:00
Kai Yang 5a6801dde7 [Core] Remove delete_creating_tasks (#12962) 2020-12-22 00:01:27 +08:00
Hao Chen 6e354690b6 [Java] Make task options serializable (#13010) 2020-12-21 23:58:43 +08:00
fangfengbin 85a4435ba0 [GCS]Fix redis store client AsyncPutWithIndex unordered bug (#13002) 2020-12-21 20:02:50 +08:00
Barak Michener c576f0b073 [ray_client] Implement a gRPC streaming logs API for the client (#13001) 2020-12-20 19:35:34 -08:00
fangfengbin 4caa6c6d78 [GCS]GCS resource manager remove cluster_resources_ (#12972) 2020-12-21 11:00:25 +08:00
Sven Mika b2bcab711d [RLlib] Attention Nets: tf (#12753) 2020-12-20 20:22:32 -05:00
Barak Michener e715ade2d1 Support retrieval of named actor handles (#13000)
Change-Id: I05d31c9c67943d2a0230782cbdaa98341584cbc7
2020-12-20 16:34:50 -08:00
Barak Michener 80f6dd16b2 [ray_client] Implement optional arguments to ray.remote() and f.options() (#12985) 2020-12-20 15:43:48 -08:00
Ameer Haj Ali 11f34f72d8 [autoscaler] Do not count head node with min_workers constraint. (#12980) 2020-12-20 14:54:46 -08:00
Barak Michener 7ab9164f1b [ray_client] Integrate with test_basic, test_basic_2 and test_actor (#12964) 2020-12-20 14:54:18 -08:00
Philipp Moritz bf6577c8f4 Switch debugger to sockets and support unicode (#13004) 2020-12-20 12:10:28 -08:00
Ian Rodney d6e243ad46 [serve] Refactor to full control loop design (#12537) 2020-12-20 13:03:57 -06:00
Sven Mika 407a3523f3 [RLlib] eval_workers after restore not generated in Trainer due to unintuitive config handling. (#12844) 2020-12-20 09:37:31 -05:00
fangfengbin 3fab93b61b Fix scheduling_resources comment errors (#12991)
* Fix scheduling_resources comment error

* add part code

Co-authored-by: 灵洵 <fengbin.ffb@antgroup.com>
2020-12-20 20:20:07 +08:00
Richard Liaw 038a50af52 [tune] skopt fix-extra-import (#12970)
Signed-off-by: Richard Liaw <rliaw@berkeley.edu>
2020-12-20 01:01:09 -08:00
Philipp Moritz ec9ad4a56b Documentation for Ray debugger stepping (#12845) 2020-12-20 00:43:27 -08:00
Amog Kamsetty 4c63917439 [Queue] Add options and shutdown to Queue (#12932)
Co-authored-by: Richard Liaw <rliaw@berkeley.edu>
2020-12-20 00:42:21 -08:00
Amog Kamsetty 51139ed37c [SGD] Fix process group timeout units (#12477) 2020-12-19 21:46:33 -08:00
Dmitri Gekhtman 4832b39066 Suggest mounting into home. Note non-root user. (#12987) 2020-12-19 16:09:24 -08:00
Eric Liang 64c97d25d3 Enable by default new scheduler (#12735) 2020-12-19 13:22:24 -08:00
Amog Kamsetty 5d3c9c8861 [Tune] Mlflow Integration (#12840)
Co-authored-by: Kai Fricke <krfricke@users.noreply.github.com>
Co-authored-by: Richard Liaw <rliaw@berkeley.edu>
2020-12-19 00:40:02 -08:00
Eric Liang 5d987f5988 Revert "Increase the number of unique bits for actors to avoid handle collisions (#12894)" (#12988)
This reverts commit 3e492a79ec.
2020-12-18 23:51:44 -08:00
dHannasch a092433bc8 [core] Use the ConnectWithoutRetries error message (#12732) 2020-12-18 22:34:34 -08:00
SangBin Cho 9d939e6674 [Object Spilling] Implement level triggered logic to make streaming shuffle work + additional cleanup (#12773) 2020-12-18 19:31:14 -08:00
Alex Wu 404161a3ff [Autoscaler/Core] Remove autoscaler spam (#12952) 2020-12-18 18:22:45 -08:00
Kai Yang ac5ea2c13d [Java] Fix output parsing in RunManager (#12968)
* Fix output parsing in RunManager

* change log level

Co-authored-by: 灵洵 <fengbin.ffb@antgroup.com>
2020-12-18 18:22:12 -08:00