Avnish Narayan
aee008ab49
[RLlib] PPO release tests tuned and re-enabled. ( #27564 )
2022-08-08 21:04:19 +02:00
Avnish Narayan
6a31b61580
[RLlib] CQL change hparams and data reading strategy ( #27451 )
2022-08-04 18:55:32 -07:00
Avnish Narayan
55209692ee
[RLlib] Deflake MARWIL and BC and remove memory leak from torch MARWIL policy ( #27406 )
2022-08-03 16:53:12 -07:00
kourosh hakhamaneshi
bda5026428
[RLlib] Fix A2C release tests ( #27314 )
2022-08-02 10:44:52 -07:00
Jun Gong
e6e10ce4cf
[RLlib] Revert 41c9ef70. ( #27243 )
...
Why are these changes needed?
Also:
Add validation to make sure multi-gpu and micro-batch is not used together.
Update A2C learning test to hit the microbatching branch.
Minor comment updates.
2022-07-29 11:05:15 -07:00
Avnish Narayan
f5a9a44b9c
[RLlib] Revert Revert Fix apex long running test ( #26928 )
2022-07-26 15:10:25 -07:00
Avnish Narayan
a50a81a13a
Revert "[RLlib] Fix apex breakout release test performance. ( #26867 )" ( #26927 )
2022-07-23 17:27:50 +02:00
Avnish Narayan
2cfd6c2e97
[RLlib] Fix apex breakout release test performance. ( #26867 )
2022-07-23 13:53:03 +02:00
Avnish Narayan
82395c4646
[RLlib] Put learning test into own folders ( #26862 )
...
Co-authored-by: Artur Niederfahrenhorst <artur@anyscale.com>
2022-07-22 11:20:47 -07:00
Artur Niederfahrenhorst
4ce9686d94
[RLlib] Fixes MARWIL release tests ( #26586 )
2022-07-15 11:13:15 -07:00
Jun Gong
c026374acb
[RLlib] Fix the 2 failing RLlib release tests. ( #25603 )
2022-06-14 14:51:08 +02:00
Sven Mika
7c39aa5fac
[RLlib] Trainer.training_iteration -> Trainer.training_step; Iterations vs reportings: Clarification of terms. ( #25076 )
2022-06-10 17:09:18 +02:00
Rohan Potdar
a9d8da0100
[RLlib]: Doubly Robust Off-Policy Evaluation. ( #25056 )
2022-06-07 12:52:19 +02:00
Rohan Potdar
ab81c8e9ca
[RLlib]: Rename input_evaluation to off_policy_estimation_methods. ( #25107 )
2022-05-27 13:14:54 +02:00
Steven Morad
501d932449
[RLlib] SAC, RNNSAC, and CQL TrainerConfig objects ( #25059 )
2022-05-22 19:58:47 +02:00
Artur Niederfahrenhorst
fb2915d26a
[RLlib] Replay Buffer API and Ape-X. ( #24506 )
2022-05-17 13:43:49 +02:00
Sven Mika
0cd7bc4054
[RLlib] Re-establish dashboard performance tests. ( #24728 )
2022-05-16 13:13:49 +02:00
Sven Mika
70d3bfcf9c
[RLlib] Provide more time for APPO Pong release and performance tests. ( #24503 )
2022-05-05 18:19:38 +02:00
Sven Mika
b48f63113b
[RLlib] SlateQ fixes: Release learning tests wrong yaml structure + TD-error torch issue ( #24429 )
2022-05-04 13:37:14 +02:00
Sven Mika
f066180ed5
[RLlib] Deprecate timesteps_per_iteration config key (in favor of min_[sample|train]_timesteps_per_reporting. ( #24372 )
2022-05-02 12:51:14 +02:00
Sven Mika
3052193c9e
[RLlib] Fix CQL getting stuck when deprecated timesteps_per_iteration is used (use min_train_timesteps_per_reporting instead). ( #24345 )
...
Fix CQL getting stuck when deprecated timesteps_per_iteration is used (use min_train_timesteps_per_reporting instead).
CQL does not perform sampling timesteps and the deprecated timesteps_per_iteration is automatically translated into the new min_sample_timesteps_per_reporting, but should be translated (only for CQL and other purely offline RL algos) into min_train_timesteps_per_reporting.
If timesteps_per_iteration, CQL lever leaves the first iteration as it thinks it's not done yet (sample timesteps always remain at 0).
2022-04-29 21:02:34 +01:00
Avnish Narayan
fdc6e02c29
[RLlib; testing] Move num_workers to RLlib config ( #23750 )
2022-04-06 20:06:48 +02:00
Avnish Narayan
161d95c31b
[RLlib] Increase slateq workers to decrease runtime on prod ( #23609 )
2022-03-30 17:38:21 -07:00
Artur Niederfahrenhorst
9a64bd4e9b
[RLlib] Simple-Q uses training iteration fn (instead of execution_plan); ReplayBuffer API for Simple-Q ( #22842 )
2022-03-29 14:44:40 +02:00
Sven Mika
22c9c4aa39
[RLlib] Slate-Q +GPU torch bug fix. ( #23464 )
2022-03-24 17:39:33 +01:00
Avnish Narayan
9040f54060
[RLlib] Pin Gym Everywhere and turn off gpu for recsim tests ( #23452 )
2022-03-24 09:17:30 +01:00
Avnish Narayan
6c20e9d898
[RLlib] Change the slateq regression learning test with GPU to use torch only ( #23168 )
2022-03-16 09:15:59 +01:00
Sven Mika
7b687e6cd8
[RLlib] SlateQ: Add a hard-task learning test to weekly regression suite. ( #22544 )
2022-02-25 21:58:16 +01:00
Jun Gong
04dd536987
[Release tests] Disable A3C CI tests on torch for now. Also extend performance_test deadline to 3hrs. ( #22426 )
2022-02-16 13:06:09 +01:00
Sven Mika
04a5c72ea3
Revert "Revert "[RLlib] Speedup A3C up to 3x (new training_iteration function instead of execution_plan) and re-instate Pong learning test."" ( #18708 )
2022-02-10 13:44:22 +01:00
Alex Wu
b122f093c1
Revert "[RLlib] Speedup A3C up to 3x (new training_iteration function instead of execution_plan) and re-instate Pong learning test." ( #22250 )
...
Reverts ray-project/ray#22126
Breaks rllib:tests/test_io
2022-02-09 09:26:36 -08:00
Sven Mika
ac3e6ab411
[RLlib] Speedup A3C up to 3x (new training_iteration function instead of execution_plan) and re-instate Pong learning test. ( #22126 )
2022-02-08 19:04:13 +01:00
Balaji Veeramani
7f1bacc7dc
[CI] Format Python code with Black ( #21975 )
...
See #21316 and #21311 for the motivation behind these changes.
2022-01-29 18:41:57 -08:00
Jun Gong
1315293dd8
[RLlib] Fix offline RL(BC & MARWIL) weekly learning tests. ( #21643 )
2022-01-18 09:29:01 +01:00
Jun Gong
7517aefe05
[RLlib] Bring back BC and Marwil learning tests. ( #21574 )
2022-01-14 14:35:32 +01:00
Jun Gong
767f78eaf8
[RLlib] Always attach latest eval metrics. ( #21011 )
2021-12-15 11:42:53 +01:00
gjoliver
d8a61f801f
[RLlib] Create a set of performance benchmark tests to run nightly. ( #19945 )
...
* Create a core set of algorithms tests to run nightly.
* Run release tests under tf, tf2, and torch frameworks.
* Fix
* Add eager_tracing option for tf2 framework.
* make sure core tests can run in parallel.
* cql
* Report progress while running nightly/weekly tests.
* Innclude SAC in nightly lineup.
* Revert changes to learning_tests
* rebrand to performance test.
* update build_pipeline.py with new performance_tests name.
* Record stats.
* bug fix, need to populate experiments dict.
* Alphabetize yaml files.
* Allow specifying frameworks. And do not run tf2 by default.
* remove some debugging code.
* fix
* Undo testing changes.
* Do not run CQL regression for now.
* LINT.
Co-authored-by: sven1977 <svenmika1977@gmail.com>
2021-11-08 18:15:13 +01:00
gjoliver
e9f66cc394
Reduce success criteria for a few learning tests. ( #19484 )
2021-10-18 15:44:38 -07:00
Sven Mika
e6aae61487
[RLlib; testing] Fix bug in stress tests not handling >1 trials per experiment (due to grid-search in IMPALA stress tests). ( #18705 )
2021-09-20 15:31:57 +02:00
Sven Mika
ba1c489b79
[RLlib Testing] Lower --smoke-test "time_total_s" to make sure it doesn't time out. ( #18670 )
2021-09-16 18:22:23 +02:00
Sven Mika
08c09737fa
[RLlib] Fix R2D2 (torch) multi-GPU issue. ( #18550 )
2021-09-14 19:58:10 +02:00
Sven Mika
cabaa3b3c6
[RLlib Testing] Add A3C/APPO/BC/DDPPO/MARWIL/CQL/ES/ARS/TD3 to weekly learning tests. ( #18381 )
2021-09-07 11:48:41 +02:00
Sven Mika
9a8ca6a69d
[RLlib] Fix Atari learning test regressions (2 bugs) and 1 minor attention net bug. ( #18306 )
2021-09-03 13:29:57 +02:00
Sven Mika
a7670d9fab
[RLlib; Testing] Fix smoke-test settings for nightly learning_tests and stress_test; Add pybullet_envs to app-config. ( #18274 )
2021-09-01 21:46:06 +02:00
Sven Mika
8acb469b04
[RLlib; Testing] Green all RLlib nightly tests. ( #18073 )
2021-08-26 14:09:20 +02:00
Sven Mika
9883505e84
[RLlib] Add [LSTM=True + multi-GPU]-tests to nightly RLlib testing suite (for all algos supporting RNNs, except R2D2, RNNSAC, and DDPPO). ( #18017 )
2021-08-24 21:55:27 +02:00
Sven Mika
a428f10ebe
[RLlib] Add multi-GPU learning tests to nightly. ( #17778 )
2021-08-18 17:21:01 +02:00
Kai Fricke
bd2404e496
[release/rllib] fix learning test script ( #17635 )
2021-08-06 10:07:59 +01:00
Sven Mika
c9d220bcda
[RLlib] Upgrade RLlib regression test scripts to new testing tool - RLlib release logs for 1.4. ( #16080 )
2021-06-01 17:39:18 +02:00
