hiro/ray
1
0
Fork 0
mirror of https://github.com/vale981/ray synced 2025-03-06 10:31:39 -05:00
Code Issues Projects Releases Packages Wiki Activity Actions
1347 commits 227 branches 67 tags 234 MiB
Commit graph

1347 commits

This branch
This branch
All branches
Author SHA1 Message Date
Philipp Moritz
7550b628bf fix indentation for ES (#1484) 2018-01-31 17:22:38 -08:00
Robert Nishihara
7187f9fe56 Pin gym version to 0.9.5 in tests. (#1490) 2018-01-31 15:50:25 -08:00
Philipp Moritz
a3f8fa426b Start integrating new GCS APIs (#1379) 
* Start integrating new GCS calls

* fixes

* tests

* cleanup

* cleanup and valgrind fix

* update tests

* fix valgrind

* fix more valgrind

* fixes

* add separate tests for GCS

* fix linting

* update tests

* cleanup

* fix python linting

* more fixes

* fix linting

* add plasma manager callback

* add some documentation

* fix linting

* fix linting

* fixes

* update

* fix linting

* fix

* add spillback count

* fixes

* linting

* fixes

* fix linting

* fix

* fix

* fix
 2018-01-31 11:01:12 -08:00
Eric Liang
35b1d6189b
[tune] save error msg, cleanup after object checkpoints 2018-01-29 18:48:45 -08:00
Robert Nishihara
0b022c0973 Automatically increase redis maxclients if the ulimit is high enough. (#1482) 2018-01-28 12:55:38 -08:00
Robert Nishihara
4c6dae5517 Raise an exception in Jenkins tests after a timeout. (#1477) 2018-01-27 20:21:27 -08:00
Eric Liang
8c96c34bb3 [tune] fix doc typo and also make sure to clean "/" from config (#1476) 2018-01-26 21:51:07 -08:00
Robert Nishihara
3195c6aa63 Fix local scheduler crash when driver creates actor and exits. (#1474) 
* Make check failures in redis.cc more informative.

* Fix bug by calling task_table_add_task.

* Add test.
 2018-01-26 14:29:53 -08:00
Stephanie Wang
668737f383 Replace actor dummy objects with mock calls to the local scheduler (#1467) 
* Replace putting the dummy object with a call to the local scheduler

* Mark dummy objects as locally available
 2018-01-26 14:18:45 -08:00
Zongheng Yang
782b4aeb0f Document how to profile Ray using pprof. (#1464) 
* Document how to profile Ray using pprof.

* Link to profiling.rst via main doc site; fix lint

* Some changes and clarifications.
 2018-01-25 21:40:52 -08:00
Kaahan
7aa979a024 [tune] Added Population Based Training (#1355) 
Adds a Population-Based Training (as described in https://arxiv.org/abs/1711.09846) scheduler to Ray.tune. Currently mutates hyperparameters according to either a user-defined list of possible values to mutate to (necessary if hyperparameters can only be certain values ex. sgd_batch_size), or by a factor of 0.8 or 1.2.
 2018-01-25 21:38:37 -08:00
Richard Liaw
e5c4d9ea0c
[tune] Fix Trial Logging File name (#1466) 2018-01-25 17:57:40 -08:00
Richard Liaw
f3d2dc0ad4
[tune] Fix Docs (#1469) 2018-01-25 16:39:00 -08:00
Robert Nishihara
e96acc26f7 Fix MNIST downloading problems in parameter server examples. (#1457) 
* Fix MNIST downloading problems in parameter server examples.

* Improve seeding.

* Fixes.
 2018-01-25 14:14:37 -08:00
eugenevinitsky
0a01d3c71f [rllib] Mountaincar fix (#1472) 
* Fix for gym version 0.9.5.

* fixed bug in reshaper that was causing discrete spaces to fail
 2018-01-25 13:58:35 -08:00
Robert Nishihara
f6c835e4b8 Fix for gym version 0.9.5. (#1471) 2018-01-25 13:58:15 -08:00
Robert Nishihara
ab5d4a6010 Bring cloudpickle inside the repository. (#1445) 
* Bring cloudpickle version 0.5.2 inside the repo.

* Use internal copy of cloudpickle everywhere.

* Fix linting.

* Import ordering.

* Change __init__.py.

* Set pickler in serialization context.

* Don't check ray location.
 2018-01-25 11:36:37 -08:00
Eric Liang
173f1d629a
[tune] Ray Tune API cleanup (#1454) 
Remove rllib dep: trainable is now a standalone abstract class that can be easily subclassed.

Clean up hyperband: fix debug string and add an example.

Remove YAML api / ScriptRunner: this was never really used.

Move ray.init() out of run_experiments(): This provides greater flexibility and should be less confusing since there isn't an implicit init() done there. Note that this is a breaking API change for tune.
 2018-01-24 16:55:17 -08:00
Philipp Moritz
a1b01ee7fb [autoscaler] Fix confirmation (y/N) for autoscaler for Python 2 (#1450) 
* Fix autoscaler for Python 2

* fix version test

* add linting exception for raw_input in Python 3

* two spaces
 2018-01-24 14:05:42 -08:00
Richard Liaw
a7d544424c [tune] Experiment Management API (#1328) 
* init for exposing external interface

* revisions

* http server

* small

* simplify

* ui

* fixes

* test

* nit

* nit

* merge

* untested

* nits

* nit

* init tests

* tests

* more tests

* nit

* fix hyperband

* cleanup

* nits

* good stuff

* cleanup

* comments and need to test

* nit

* notebook

* testing

* test and expose server

* server_tests

* docs

* periods

* fix tests

* committing test

* fi
 2018-01-24 13:45:10 -08:00
Eric Liang
1d2a28ab07 [rllib] test all combinations of {obs_space} x {action_space} (#1449) 2018-01-24 11:03:43 -08:00
Robert Nishihara
5acc98e629 Update arrow with better dataframe serialization and get rid of custo… (#1413) 
* Update arrow with better dataframe serialization and get rid of custom dataframe serializers.

* Update plasma client API.

* Fix potential bug.

* Bug fix.

* Update arrow to use deduplicated file descriptors and mutable buffers.

* Fix tests.

* Update commit.

* Update commit.

* Update commit.

* Update commit.

* Update commit

* Update commit back to arrow codebase.'
 2018-01-24 10:03:29 -08:00
Alexey Tumanov
f1303291b4 Ray scheduler spillback plumbing + mechanism (#1362) 
* spillback mechanism and plumbing : adding spillback counter + timestamp

* linting fix

* documentation

* Fix argument name.
 2018-01-23 20:18:12 -08:00
Devin Petersohn
21a916009e Adding documentation for building from current master without cloning. (#1455) 
* Updating docs with pip way to build from source

* wording
 2018-01-23 14:47:30 -08:00
Stephanie Wang
fb4d991732
Add actor reconstruction limitations to documentation (#1452) 
* Add limitations to actor fault tolerance

* rename heading
 2018-01-23 13:40:50 -08:00
Roy Fox
4b0ef5eb2c [rllib] Behavior Cloning (#1400) 
* Behavior Cloning

* episode_reward_mean -> mean_loss

* removing vestigial code

* punctuation

* unnecessary

* Behavior Cloning

* Behavior Cloning

* Update __init__.py
 2018-01-23 10:50:45 -08:00
Eric Liang
ee36effd8e
[rllib] Add n-step Q learning for DQN (#1439) 
* n-step

* add sample adjustm

* Oops

* fix nstep

* metric adjustment

* Sat Jan 20 23:30:34 PST 2018

* Sun Jan 21 16:40:46 PST 2018

* Mon Jan 22 22:24:57 PST 2018
 2018-01-23 10:31:19 -08:00
Robert Nishihara
f32c0c8ec1 Move calls to ray.worker.cleanup into tearDown part of tests for isolation. (#1433) 2018-01-22 22:54:56 -08:00
Melih Elibol
4b1c8be4fe Fix setting log-level to debug. (#1432) 2018-01-21 21:51:05 -08:00
Devin Petersohn
4aca016bff Adding series and a way to validate our API. (#1435) 
* Adding series and a way to validate our API.

* Moving partitions into protected status
 2018-01-21 19:20:54 -08:00
Stephanie Wang
74718efa73
Nondeterministic reconstruction for actors (#1344) 
* Add failing unit test for nondeterministic reconstruction

* Retry scheduling actor tasks if reassigned to local scheduler

* Update execution edges asynchronously upon dispatch for nondeterministic reconstruction

* Fix bug for updating checkpoint task execution dependencies

* Update comments for deterministic reconstruction

* cleanup

* Add (and skip) failing test case for nondeterministic reconstruction

* Suppress test output
 2018-01-21 13:44:13 -08:00
Peter Schafhalter
83949a533b [autoscaler] Increased head and worker storage to 25 GiB (#1401) 
* Increased head and worker storage to 25 GiB

* Update example.yaml
 2018-01-21 13:09:29 -08:00
Richard Liaw
04a50aa9ae
[tune] Standardize Ray Tune on documentation (#1448) 2018-01-21 12:07:15 -08:00
Eric Liang
a2b190e65b Fix occasional task timeline failure to get task ids (#1442) 2018-01-21 12:04:44 -08:00
Eric Liang
424bd7f74d
[rllib] improve custom env docs (#1447) 
* env docs

* add env

* update env

* Fri Jan 19 18:55:34 PST 2018
 2018-01-19 21:36:18 -08:00
Robert Nishihara
d7dfb16cc8 Add some development tips to documentation. (#1426) 
* Add some development tips to documentation.

* Add more tips.

* Add permission denied help.
 2018-01-19 16:16:45 -08:00
Robert Nishihara
ed603fc5e9 Add link to github from documentation. (#1425) 
* Add link to github from documentation.

* Add fork me on github banner.
 2018-01-19 10:14:34 -08:00
Eric Liang
e216766bbc [rllib] Update docs with api and components overview figures (#1443) 2018-01-19 10:08:45 -08:00
eugenevinitsky
37076a9ff8 Multiagent model using concatenated observations (#1416) 
* working multi action distribution and multiagent model

* currently working but the splits arent done in the right place

* added shared models

* added categorical support and mountain car example

* now compatible with generalized advantage estimation

* working multiagent code with discrete and continuous example

* moved reshaper to utils

* code review changes made, ppo action placeholder moved to model catalog, all multiagent code moved out of fcnet

* added examples in

* added PEP8 compliance

* examples are mostly pep8 compliant

* removed all flake errors

* added examples to jenkins tests

* fixed custom options bug

* added lines to let docker file find multiagent tests

* shortened example run length

* corrected nits

* fixed flake errors
 2018-01-18 19:51:31 -08:00
Peter Schafhalter
215d526e0d Load evaluation configuration from checkpoint (#1392) 2018-01-17 10:51:33 -08:00
Eric Liang
b8811cbe34
[autoscaling] increase connect timeout, boto retries, and check subnet conf (#1422) 
* some autoscaling config tweaks

* Sun Jan 14 13:56:55 PST 2018

* Mon Jan 15 14:21:09 PST 2018

* increase backoff

* Mon Jan 15 14:40:47 PST 2018

* check boto version
 2018-01-16 16:11:09 -08:00
Robert Nishihara
eac11c252c Update wheel in autoscaler example. (#1408) 2018-01-13 01:06:23 -08:00
Yaroslav Bulatov
78fb3c5ed9 [autoscaler] Fix ValueError: Missing required config keyavailability_zoneof type str 2018-01-13 00:59:15 -08:00
Richard Liaw
d4592382a4
[tune][minor] Fixes (#1383) 2018-01-11 18:14:20 -08:00
Philipp Moritz
1290072764 [rllib] Expose PPO evaluator resource requirements (#1391) 2018-01-11 11:09:01 -08:00
Philipp Moritz
44792530a9 fix autoscaler test (#1411) 2018-01-10 13:18:34 -08:00
Eric Liang
5a2f85048d [rllib] Fix incorrect documentation on how to use custom models #1405 2018-01-09 18:09:05 -08:00
Peter Schafhalter
a59a9e20af Added option for availability zone (#1393) 2018-01-09 13:49:47 -08:00
Devin Petersohn
112ef07563 Adding all DataFrame methods with NotImplementedErrors (#1403) 
* Adding all DataFrame methods with NotImplementedErrors

* Moving dataframe creation into function call
 2018-01-07 12:00:16 -08:00
Robert Nishihara
1e0dfca2dc Remove pyarrow version check. (#1394) 2018-01-06 22:42:55 -08:00