Sven Mika
|
cf7a09daa2
[RLlib] Provide more constants for common result dict keys, e.g. `EPISODE_RETURN_MEAN`. (#45330)
|
5 月之前 |
Sven Mika
|
611e7d5509
[RLlib] Metrics do-over 02: Use new MetricsLogger in `PPO.training_step()` and `EnvRunner.get_metrics()`. (#44888)
|
6 月之前 |
Sven Mika
|
4b47995420
[RLlib] Evaluation do-over: Make parallel evaluation to training the default behavior and deprecate async eval option. (#43787)
|
7 月之前 |
Sven Mika
|
f8e59cba73
[RLlib] Do-over of release tests in light of rllib_contrib AND new- vs old API stack. (#43278)
|
7 月之前 |
Sven Mika
|
0f62ccc6ee
[RLlib] Clean up some deprecation messages (they shouldn't be there) and make others `error=True` (from `error=False`) (#38555)
|
1 年之前 |
Sven Mika
|
b0045799e3
[RLlib] DreamerV3: Make 200M (XL model) work; mixed float16 option (#38461)
|
1 年之前 |
Sven Mika
|
8c055af084
[RLlib] DreamerV3: Add CI Testing. (#37979)
|
1 年之前 |
Sven Mika
|
656fe0703c
Revert revert [RLlib] DreamerV3 Main Algo. (#36571)
|
1 年之前 |
Sven Mika
|
a3ec4a936e
[RLlib] Enable `eager_tracing=True` by default. (#36556)
|
1 年之前 |
Kai Fricke
|
42e06e3948
Revert "[RLlib] DreamerV3: Main algo code and required changes to some RLlib APIs (RolloutWorker). (#35386)" (#36564)
|
1 年之前 |
Sven Mika
|
8290bd112c
[RLlib] DreamerV3: Main algo code and required changes to some RLlib APIs (RolloutWorker). (#35386)
|
1 年之前 |
Sven Mika
|
61f2dc1b05
[RLlib] Enhance `run_regression_tests.py`: Allow overriding `--env` and `--framework` on command line. (#35985)
|
1 年之前 |
Sven Mika
|
e399fb8037
[RLlib] APPO+new-stack (Atari benchmark) - Preparatory PR 02. (#34777)
|
1 年之前 |
Sven Mika
|
df76ac7975
[RLlib] Multi-agent test and support envs that sometimes return empty obs-dicts (no agent requesting an action). (#30413)
|
1 年之前 |
Sven Mika
|
5883e9ae41
[RLlib] Enable .py files with AlgorithmConfig objects in them to be run by our `run_regression_test.py` script. (#29974)
|
1 年之前 |
Kai Fricke
|
6b9a56d28e
[ci/docker/air] Update ML/DL dependencies to latest releases / Remove Py3.6 Docker images (#28808)
|
2 年之前 |
Sven Mika
|
432f023642
[RLlib] Deprecate `AlgorithmConfig.framework("tfe")`: Use `tf2` instead. (#29755)
|
2 年之前 |
kourosh hakhamaneshi
|
68a35fceef
[RLlib] unify target update method (#28135)
|
2 年之前 |
Avnish Narayan
|
a322ac463c
[RLlib] Make JSONReader default, users will have to use the DatasetReader for any speedups. (#26541)
|
2 年之前 |
kourosh hakhamaneshi
|
c38a29573f
[RLlib] Removed deprecated code with error=True (#23916)
|
2 年之前 |
Sven Mika
|
6522935291
[RLlib] Slate-Q tf implementation and tests/benchmarks. (#22389)
|
2 年之前 |
Avnish Narayan
|
0d2ba41e41
[RLlib] [CI] Deflake longer running RLlib learning tests for off policy algorithms. Fix seeding issue in TransformedAction Environments (#21685)
|
2 年之前 |
Balaji Veeramani
|
7f1bacc7dc
[CI] Format Python code with Black (#21975)
|
2 年之前 |
Sven Mika
|
abd3bef63b
[RLlib] QMIX better defaults + added to CI learning tests (#21332)
|
2 年之前 |
Sven Mika
|
2d24ef0d32
[RLlib] Add all simple learning tests as `framework=tf2`. (#19273)
|
3 年之前 |
Carlo Grisetti
|
5cee8a1985
[release tests] Switch from yaml.load to yaml.safe_load (#19365)
|
3 年之前 |
Sven Mika
|
8a72824c63
[RLlib Testig] Split and unflake more CI tests (make sure all jobs are < 30min). (#18591)
|
3 年之前 |
Sven Mika
|
cabaa3b3c6
[RLlib Testing] Add A3C/APPO/BC/DDPPO/MARWIL/CQL/ES/ARS/TD3 to weekly learning tests. (#18381)
|
3 年之前 |
Sven Mika
|
e3e6ed7aaa
[RLlib] Issues 17844, 18034: Fix n-step > 1 bug. (#18358)
|
3 年之前 |
Sven Mika
|
7f2b3c0824
[RLlib] Issue 17667: CQL-torch + GPU not working (due to simple_optimizer=False; must use simple optimizer!). (#17742)
|
3 年之前 |