hiro/ray
1
0
Fork 0
mirror of https://github.com/vale981/ray synced 2025-03-05 18:11:42 -05:00
Code Issues Projects Releases Packages Wiki Activity Actions
2423 commits 227 branches 67 tags 234 MiB
Commit graph

2423 commits

This branch
This branch
All branches
Author SHA1 Message Date
Si-Yuan
48139cf861 Migrate Python C extension to Cython (#3541) 2019-01-24 09:17:14 -08:00
Yuhong Guo
c1a52b1c86 Remove duplicated code in RayConfig (#3831) 2019-01-24 17:04:10 +08:00
Hao Chen
bfcf254e52 Fix: do not treat actor task as failed if the actor will be reconstructed (#3736) 2019-01-23 23:28:44 -08:00
Eric Liang
04ec47cbd4
[rllib] annotate public vs developer vs private APIs (#3808) 2019-01-23 21:27:26 -08:00
Robert Nishihara
01e18b47f4 Direct people to stackoverflow for questions about usage. (#3830) 
* Direct people to stackoverflow for questions about usage.

* Improve wording
 2019-01-23 13:30:02 -08:00
Wang Qing
dcb744518e Implement actor dummy object gc in java (#3822) 
* Add dummy object gc in java

* Fix

* Address comments.

* Refine

* Address comments.
 2019-01-23 11:56:25 -08:00
Wang Qing
816406ea3d [Java] Fix setCurrentTask() in multi threading (#3821) 2019-01-23 20:45:30 +08:00
Robert Nishihara
0b1608a546 Factor out code for starting new processes and test plasma store in valgrind. (#3824) 
* Factor out starting Ray processes.

* Detect flags through environment variables.

* Return ProcessInfo from start_ray_process.

* Print valgrind errors at exit.

* Test valgrind in travis.

* Some valgrind fixes.

* Undo raylet monitor change.

* Only test plasma store in valgrind.
 2019-01-22 14:59:11 -08:00
Eric Liang
f0e6523323
[rllib] Don't call reset() unless necessary for multi-agent envs 2019-01-20 15:00:18 -08:00
Philipp Moritz
0dad4e6a25 Build Raylet with Bazel (#3806) 2019-01-20 12:16:47 -08:00
Eric Liang
aad48ee5a5 [tune] Fully deprecate raw function literals in Tune (#3788) 
Related: https://github.com/ray-project/ray/issues/3785
 2019-01-19 17:09:36 -08:00
Michael Luo
16f7ca45e4 Appo (#3779) 
* Deleted old fork, updated new ray and moved PPO-impala to APPO in ppo folder

* Deleted unneccesary vtrace.py file

* Update pong-impala.yaml

* Cleaned PPO Code

* Update pong-impala.yaml

* Update pong-impala.yaml

* wip

* new ifle

* refactor

* add vtrace off option

* revert

* support any space

* docs

* fix comment

* remove kl

* Update cartpole-appo-vtrace.yaml
 2019-01-18 13:40:26 -08:00
Philipp Moritz
931e6a2fc3 Fix compilation error on ARM. (#3800) 2019-01-18 00:25:16 -08:00
Robert Nishihara
9af5a62e05 Give better error for old-style actor classes. (#3793) 2019-01-17 19:05:04 -08:00
Richard Liaw
0537508106 Bump strings for 0.6.2 (#3801) 2019-01-17 19:03:27 -08:00
Si-Yuan
16a3b99d8d Get rid of Arrow test utils (#3734) 
* convert code to proper C++

* revert changes to "id.h" because #3765 has been merged.

* revert changes to Python bindings because they will be removed in #3541

* remove dependencies of Arrow logging

* revert changes to Arrow logging

* lint
 2019-01-17 18:35:41 -08:00
Jones Wong
319c1340cb [rllib] Develop MARWIL (#3635) 
*  add marvil policy graph

*  fix typo

*  add offline optimizer and enable running marwil

*  fix loss function

*  add maintaining the moving average of advantage norm

*  use sync replay optimizer for unifying

*  remove offline optimizer and use sync replay optimizer

*  format by yapf

*  add imitation learning objective

*  fix according to eric's review

*  format by yapf

* revise

* add test data

* marwil
 2019-01-16 19:00:43 -08:00
Hao Chen
d1840bc7a9 Simplify RayConfig (#3714) 2019-01-16 16:43:26 -08:00
Richard Liaw
75ac016e2b Bump version (#3787) 2019-01-16 11:40:54 -08:00
Richard Liaw
fa99fda2b4
Application Stress Tests (#3612) 2019-01-16 02:05:16 -08:00
Richard Liaw
c28e6d41f5
[tune] Avoid overwriting checkpoint file (#3781) 2019-01-16 02:03:16 -08:00
ggdupont
a237b4a6a1 [Java] Fix package jaxb not exist when JDK11 (#3738) 2019-01-16 14:15:00 +08:00
Philipp Moritz
3b39066c15 Fix pandas 0.22 incompatibility by upgrading Arrow (#3786) 2019-01-15 21:17:32 -08:00
Eric Liang
401e656b95 [rllib] Sync filters at end of iteration not start; hierarchical docs (#3769) 2019-01-15 16:25:25 -08:00
Richard Liaw
3918934dfd
[tune] Cross-Node Recovery (#3725) 
Augments trial restore to also check if the runner is at the same
location. If not, the checkpoint files are pushed onto the new location.
 2019-01-15 10:37:28 -08:00
Si-Yuan
a5df8e3532 minor fix (#3770) 2019-01-14 13:52:51 -08:00
Tianming Xu
0b8008f41c remove RAY_CHECK around wait_state.remaining.erase (#3745) 2019-01-14 10:32:31 -08:00
Philipp Moritz
02bdaf221d Update arrow to include https://github.com/apache/arrow/pull/3392 (#3765) 
* update arrow to include https://github.com/apache/arrow/pull/3392

* add appropriate includes

* update
 2019-01-14 19:20:26 +08:00
Wang Qing
3cf59855af [Java] Replace junit with testNG (#3768) 2019-01-14 17:49:17 +08:00
Robert Nishihara
19908c01b8 Use environment markers to only install faulthandler in Python < 3.3. (#3764) 2019-01-14 15:55:59 +08:00
Hao Chen
1bb20badec [Java] Fix bug when actor creation task fails (#3740) 
* [Java] Fix bug when actor creation task fails

* remove imports
 2019-01-14 11:09:15 +08:00
Robert Nishihara
27c20a41a9 Update stress tests. (#3614) 
Starts clusters for testing and has a fallback to kill the cluster if the command fails.

The results are then printed at the end of test.
 2019-01-13 17:08:51 -08:00
Eugene Vinitsky
a5d1f03515 [rllib] fix for rollout of lstm policies (#3643) 
* fix for lstm policies

* added call to local evaluator

* Update python/ray/rllib/rollout.py

Co-Authored-By: eugenevinitsky <eugenevinitsky@users.noreply.github.com>

* Update rollout.py

* Update rollout.py
 2019-01-13 15:54:23 -08:00
Philipp Moritz
00e9f8d870 Fix pyarrow version (#3760) 2019-01-13 14:28:23 -08:00
jhpenger
3adffe6a4e [docs] Add example showing how to use Ray on Kubernetes. (#3126) 
Closes #1353.
 2019-01-13 13:56:47 -08:00
Wang Qing
8674606e26 Support to auto-generate Java files from flatbuffer (#3749) 
* auto gen flatbuffers for Java

* Add auto_gen_tool.py

* Refine

* Add a comment

* address comments.

* Address comments.

* Addressed

* Refine

* Address comments

* Fix typo

* Add exception

* Address comments.

* Refine

* Fix lint

* Fix

* Fix lint and address comment.

* Fix lint error
 2019-01-13 11:39:23 -08:00
Yuhong Guo
d2cf8561f2 Refactor code about ray.ObjectID. (#3674) 
* Refactor code about ray.ObjectID.

* remove from_random and use nil_id instead of constructor

* remove id() in hash

* Lint and fix

* Change driver id to ObjectID

* Replace binary_to_hex(ObjectID.id()) to ObjectID.hex()
 2019-01-13 01:47:29 -08:00
Eric Liang
c4b058739b Remove redundant error message (#3761) 2019-01-12 22:22:41 -08:00
Richard Liaw
bdeeacc70f
[autoscaler] RecoverUnhealthyWorker mitigation (#3699) 
Increases number of retries for RecoverUnhealthyWorkers

Closes #3435.
 2019-01-12 14:06:53 -08:00
Robert Nishihara
1480f309c3 [doc] Replace runtest.py with mini_test.py in documentation. (#3750) 
Rename `xray_test.py` to `mini_test.py` and use that in the documentation. Right now we suggest that people run `runtest.py`, but that often doesn't succeed and takes too long.
 2019-01-12 14:05:28 -08:00
James Casbon
528bb3afd9 gcp allow manual network configuration (#3748) 2019-01-12 14:02:20 -08:00
Robert Nishihara
fbea1ece2e Clear new actor handle list after submitting task. (#3755) 2019-01-12 23:25:40 +08:00
Wang Qing
0a556dc0b5 Refine redis client (#3758) 2019-01-12 23:01:48 +08:00
Wang Qing
a0cf8ee5a8 Refine Java worker code (#3735) 2019-01-12 22:45:33 +08:00
Robert Nishihara
8723d6b061 Define a Node class to manage Ray processes. (#3733) 
* Implement Node class and move most of services.py into it.

* Wait for nodes as they are added to the cluster.

* Fix Redis authentication bug.

* Fix bug in client table ordering.

* Address comments.

* Kill raylet before plasma store in test.

* Minor
 2019-01-11 22:30:38 -08:00
Wang Qing
fa2bfa6d76 Fix some small code quality issues. (#3719) 2019-01-11 15:24:49 +08:00
Stephanie Wang
cc5ecd71c5 [autoscaler] Add kill and get IP commands to CLI for testing (#3731) 
## What do these changes do?

Adds 2 commands to the CLI that take in an autoscaler config:
1. Kill a random ray node in the cluster.
2. Get all the worker node IP addresses.

These commands are both for testing and are not recommended for normal use.

## Related issue number
Closes #3685.
 2019-01-10 22:06:57 -08:00
Richard Liaw
574f0b73bc
[tune] Fix Trial Serialization (#3743) 2019-01-10 19:26:10 -08:00
Hao Chen
597abb24ea Refine multi-threading support (#3672) 
* [Python] refine multi-threading support

fix

* [java] refine multithreading code

fix java

* format
 2019-01-10 13:58:11 -08:00
Eric Liang
71243203a4
[rllib] Fix KeyError: 'kl' in multiagent ppo training 2019-01-09 19:33:07 -08:00