Zhijun Fu
ea9376c9ce
Fix flaky core worker tests because of race condition in gcs client subscription ( #5735 )
2019-09-24 22:47:38 +08:00
Kai Yang
c580955840
[Java] Fix some potential bugs about Ray.shutdown() ( #5693 )
2019-09-24 10:44:17 +08:00
Ujval Misra
a4659a8f8b
[tune] Add support for function-based stopping condition ( #5754 )
2019-09-23 18:39:00 -07:00
Mitchell Stern
b03147e7bf
Update call to py-spy to conform to new API ( #5758 )
2019-09-23 14:52:23 -07:00
Eric Liang
249ca2cf9e
[rllib] add blog posts to examples list ( #5762 )
...
* add blog post
* remove
* link
2019-09-23 10:42:21 -07:00
Edward Oakes
61e5d674be
Push driver task in core worker ( #5752 )
2019-09-23 10:53:55 -05:00
Edward Oakes
62bc30c1cf
Validate redis address parameters ( #5746 )
...
* Validate redis address params
* Fix comment
* Add check
2019-09-23 10:52:34 -05:00
Mitchell Stern
98dcc1d440
[Dashboard] Add initial version of new dashboard ( #5730 )
2019-09-23 08:50:40 -07:00
Eric Liang
56ab9a00bb
[autoscaler] cache stopped nodes, no screen on attach ( #5741 )
2019-09-22 17:30:35 -07:00
Philipp Moritz
5f5873b182
[Projects] Start multiple sessions via session start ( #5740 )
2019-09-22 01:36:23 -07:00
Robert Nishihara
1cfadf032e
Properly test Python wheels in Travis. ( #5749 )
2019-09-21 18:03:10 -07:00
Richard Liaw
e00071721a
[tune] tf2.0 testing and supporting callables ( #5738 )
2019-09-21 17:01:14 -07:00
Robert Nishihara
c91a37f622
Set redis password in slurm deployment documentation. ( #5747 )
2019-09-21 15:33:15 -07:00
Hersh Godse
d17b35494d
[tune] Save/Restore for Suggestion Algs ( #5719 )
2019-09-21 11:11:57 -07:00
Vince Jankovics
7e214fd95e
[tune] TensorBoard HParams for TF2.0 ( #5678 )
2019-09-21 11:06:34 -07:00
Kilian Batzner
79b9c70ad6
Add local_tf_session_args to unknown subkeys whitelist ( #5742 )
...
* Add local_tf_session_args to unknown subkeys whitelist
* Remove trailing whitespace
2019-09-20 10:32:49 -07:00
Eric Liang
fb3b232c0e
[rllib] Properly flatten 2-d observations as input to FCnet ( #5733 )
2019-09-19 12:10:31 -07:00
Matthew A. Wright
3131e1742d
[rllib] Qmix off by 1 in double Q calculation ( #5731 )
...
* Qmix fix.
-Current version of double Q learning is incorrect; it selects actions
at timestep t instead of t+1 when computing the t+1 Q value.
* Allow extra obs dict keys
* Move Q-value-computing replay code to own function
* Run the autoformatter
* use better terms in comments ("policy" network instead of "live" network)
2019-09-18 18:12:30 -07:00
gehring
8903bcd0c3
[rllib] Tracing for eager tensorflow policies with tf.function ( #5705 )
...
* Added tracing of eager policies with `tf.function`
* lint
* add config option
* add docs
* wip
* tracing now works with a3c
* typo
* none
* file doc
* returns
* syntax error
* syntax error
2019-09-17 01:44:20 -07:00
Robert Nishihara
f74aaf2619
Add more links for getting involved.git status ( #5708 )
2019-09-16 20:26:03 -07:00
Philipp Moritz
a6dd794818
[Projects] Fix template path ( #5716 )
2019-09-16 19:58:54 -07:00
Philipp Moritz
b1aadd863b
Fix project templates in wheel ( #5714 )
2019-09-16 15:21:59 -07:00
Philipp Moritz
e4e1a57ca5
[Projects] Allow named sessions ( #5706 )
2019-09-16 13:00:46 -07:00
Philipp Moritz
f4deecb5ab
Fix travis error in direct_actor_transport.cc ( #5710 )
2019-09-15 22:19:20 -07:00
Eric Liang
4bf7de084d
Speed up TaskSpecification copy ( #5709 )
2019-09-15 19:57:34 -07:00
Richard Liaw
2b2eb4debb
[tune] Checkpoint and Sync at end ( #5699 )
2019-09-15 15:58:58 -07:00
Robert Nishihara
baac370099
Deprecate old global state API. ( #5484 )
...
* Deprecate old global state API.
* Remove unnecessary returns.
2019-09-15 09:13:15 -07:00
Eric Liang
09968a3c55
[revert] Disable monitor error logging to stdout #5692
2019-09-14 22:32:48 -07:00
Eric Liang
4979b8c4d9
Ordered execution of tasks per actor handle ( #5664 )
2019-09-14 22:31:33 -07:00
Edward Oakes
a8888c5ff4
[flaky test] Fix test_calling_start_ray_head ( #5644 )
2019-09-14 22:27:45 -07:00
Robert Nishihara
74a34b736d
Call ray.put in ray.init() to speed up first object store access. ( #5685 )
2019-09-14 21:27:32 -07:00
Simon Mo
1560ace65a
Use set comprehensions ( #5707 )
2019-09-14 15:44:25 -07:00
Edward Oakes
a5d7de6aaf
[core worker] Python core worker normal task submission ( #5566 )
2019-09-14 13:02:53 -07:00
Simon Mo
5f88823c49
[Serve] Rewrite Ray.Serve From Scratch ( #5562 )
...
* Commit and format files
* address stylistic concerns
* Replcae "Usage" by "Example" in doc
* Rename srv to serve
* Add serve to CI process; Fix 3.5 compat
* Improve determine_tests_to_run.py
* Quick cosmetic for determien_tests
* Address comments
* Address comments
* Address comment
* Fix typos and grammar
Co-Authored-By: Edward Oakes <ed.nmi.oakes@gmail.com>
* Update python/ray/experimental/serve/global_state.py
Co-Authored-By: Edward Oakes <ed.nmi.oakes@gmail.com>
* Use __init__ for Query and WorkIntent class
* Remove dataclasses dependency
* Rename oid to object_id for clarity
* Rename produce->enqueue_request, consume->dequeue_request
* Address last round of comment
2019-09-13 21:36:56 -07:00
Si-Yuan
4c964c0941
Initial implementation for pickle5 support ( #5611 )
2019-09-13 17:54:14 -07:00
Simon Mo
fc9f03cd96
Fix queue actor init in setup_queue_actor fixture ( #5676 )
2019-09-13 12:35:44 -07:00
Eric Liang
3ed18d0b59
Fix edge case in autoscaler with poor bin packing ( #5702 )
...
* fix edge case
* fix for general case
2019-09-13 11:46:10 -07:00
Stephanie Wang
1d4a11a433
Only use git repo if .git exists ( #5701 )
2019-09-13 11:34:34 -07:00
Edward Oakes
07c4c6367a
[core worker] Python core worker object interface ( #5272 )
2019-09-12 23:07:46 -07:00
Kai Yang
1b880191b0
Replace NotImplementedException with UnsupportedOperationException ( #5694 )
2019-09-12 00:40:26 -07:00
Edward Oakes
ee5db5b67f
Raise error if space in redis password ( #5673 )
2019-09-11 20:58:39 -07:00
Edward Oakes
0bf79cfbde
Properly short circuit core worker Get() on exception ( #5672 )
2019-09-11 18:38:14 -07:00
Ashwinee Panda
946ebfaa3c
[rllib] Validate that entropy coeff is not an integer ( #5687 )
...
* Validate that entropy coeff is not an integer
Passing an integer value for entropy coeff such as 0 raises an error somewhere inside the TF policy graph, so this checks to make sure the entropy coeff is a float.
* Cast to float instead
Also move this check after the negative value check
2019-09-11 14:35:42 -07:00
Eric Liang
faeaa34bdd
Deflake cluster heartbeat test ( #5552 )
2019-09-11 12:26:04 -07:00
Eric Liang
bc6a95deb0
[rllib] Eager execution for centralized critic example, fix simple optimizer for multiagent ( #5683 )
2019-09-11 12:15:34 -07:00
Eric Liang
2fdefe19b7
Take into account queue length in autoscaling ( #5684 )
2019-09-11 11:31:35 -07:00
Philipp Moritz
9ce6dd9b88
[Projects] Add "session execute" ( #5681 )
2019-09-11 00:50:05 -07:00
Hersh Godse
336aef1774
[tune] Save and Restore for bayesopt ( #5623 )
2019-09-10 13:11:59 -07:00
Robert Nishihara
4d16677a68
Fix PyPI version in readme. ( #5662 )
2019-09-09 19:54:57 -07:00
Simon Mo
147e7d46ec
[Flaky tests] FIx test fork ( #5671 )
...
* Start testing test_fork
Maybe queue actor takes too long to initialize, that's why we are
seeing "Many python processes started" since most of the python
tasks are blocked on ray.get
* Add a comment
2019-09-09 19:21:20 -07:00
