Commits

Commits on Aug 1, 2024

[RLlib] MARWIL RLModule. (ray-project#44970 )
simonsays1980
committedAug 1, 2024
[RLlib; Offline RL] Offline recording new api stack. (ray-project#46818 )
simonsays1980
committedAug 1, 2024

Commits on Jul 25, 2024

[RLlib; Offline RL] Make data pipeline better configurable and tuneable for users. (ray-project#46777 )
simonsays1980
committedJul 25, 2024

Commits on Jul 23, 2024

[RLlib, Offline RL] Add user-defined schemas for data loading. (ray-project#46738 )
simonsays1980
committedJul 23, 2024

Commits on Jul 22, 2024

[RLlib] Initial design for Ray-Data based offline RL Algos (on new API stack). (ray-project#44969 )
simonsays1980
committedJul 22, 2024

Commits on Jul 17, 2024

[RLlib] Single-GPU, multi-CPU (Learner), multi-GPU tests for DQN/SAC single-/multi-agent setups in new API stack. (ray-project#46153 )
simonsays1980
committedJul 17, 2024

Commits on Jul 9, 2024

[RLlib] Enable multi-learner setup for hybrid stack BC. (ray-project#46436 )
simonsays1980
committedJul 9, 2024

Commits on Jul 8, 2024

[RLlib] Enable complex action spaces with stateful modules. (ray-project#46468 )
simonsays1980
committedJul 8, 2024

Commits on Jul 4, 2024

[RLlib] Moving sampling coordination for batch_mode=complete_episodes to synchronous_parallel_sample. (ray-project#46321 )
simonsays1980
committedJul 4, 2024

Commits on Jun 25, 2024

[RLlib] Add env- and agent_steps to custom evaluation function (unifying APIs). (ray-project#45652 )
simonsays1980
committedJun 25, 2024

Commits on Jun 21, 2024

[RLlib] Action masking example for new API stack. (ray-project#46146 )
simonsays1980
committedJun 21, 2024

Commits on Jun 14, 2024

[RLlib] Cleanup examples folder: Cleaned-up PyFlyt example. (ray-project#45956 )
simonsays1980
committedJun 14, 2024
[RLlib] Prioritized multi-agent episode replay buffer. (ray-project#45576 )
simonsays1980
committedJun 14, 2024

Commits on Jun 5, 2024

[RLlib] Fix calling of callback on_episode_created to conform to docstring (after reset). (ray-project#45651 )
simonsays1980
committedJun 5, 2024

Commits on May 31, 2024

[RLlib] Fix bug: Target nets are not synched with main nets in SAC. (ray-project#45614 )
simonsays1980
committedMay 31, 2024

Commits on May 28, 2024

[RLlib] EnvRunners fix min/max-metrics window sizes (from inf to config.metrics_num_episodes_for_smoothing). (ray-project#45575 )
simonsays1980
committedMay 28, 2024

Commits on May 24, 2024

Commits on May 22, 2024

[RLlib] Fix stateful module errors with inference only mode. (ray-project#45465 )
simonsays1980
committedMay 22, 2024

Commits on May 8, 2024

Commits on May 6, 2024

[RLlib] Fix overflow error b/c of int division in Filter updates. (ray-project#45154 )
simonsays1980
committedMay 6, 2024

Commits on May 3, 2024

[RLlib] - Fix APPO RLModule inference-only problems. (ray-project#45111 )
simonsays1980
committedMay 3, 2024

Commits on May 2, 2024

Commits on Apr 30, 2024

Commits on Apr 29, 2024

[RLlib] Fix CI test: examples/checkpoints/checkpoint_by_custom_criteria.py. (ray-project#45030 )
simonsays1980
committedApr 29, 2024

Commits on Apr 25, 2024

[RLlib] Implementation of MultiAgentEpisodeReplayBuffer for the new API stack. (ray-project#44450 )
simonsays1980
committedApr 25, 2024

Commits on Apr 23, 2024

[RLlib] "Light-weighted" RLModules for use in inference-only or exploration-only settings. (ray-project#44797 )
simonsays1980
committedApr 23, 2024

Commits on Apr 22, 2024

[RLlib] Bug fix: 'train_batch_size_per_learner' could not be used in dependent Tune search space. (ray-project#44868 )
simonsays1980
committedApr 22, 2024

Commits on Apr 17, 2024

[RLlib, Tune, AIR] Support of images and video logging in wandb. (ray-project#43356 )
simonsays1980
committedApr 17, 2024

Commits on Apr 16, 2024

[RLlib] Add example: Pre-train an RLModule single-agent, then bring checkpoint into multi-agent setup and continue training. (ray-project#44674 )
simonsays1980
committedApr 16, 2024

Commits on Apr 15, 2024

[RLlib] Better support and separation-of-concerns for model_config_dict in new API stack. (ray-project#44263 )
simonsays1980
committedApr 15, 2024