![]() Fixed error when importing models which use the ResNet encoder.Fixed an issue where the histogram stats would not be reported correctly in TensorBoard.The calculation of the target entropy of SAC with continuous actions was incorrect and has been fixed.Fixed an error where sensors would not reset properly before collecting the last observation at the end of an episode.Updated the documentation with a note saying that GridSensor does not work in 2D environments.RigidBodySensorComponent now displays a warning if it's used in a way that won't generate useful observations.Colab notebooks illustrating the use of the Python API were added to the repository.Added a fully connected visual encoder for environments with very small image inputs.Locked cattrs dependency version to 1.6.Added stacking option in Editor for VectorSensorComponent.Fixed NullReferenceException when adding Behavior Parameters with no Agent.Release Notes Minor Changes -agents / (C#) Harden user PII protection logic and extend TrainingAnalytics to expose detailed configuration parameters.Fixed a bug in VAIL where the variational bottleneck was not properly passing gradients ( #5546).Fixed failing tests for gym-unity due to gym 0.20.0 release ( #5540).Fixed conflicting CLI and yaml options regarding resume & initialize_from ( #5495).Fixed wrong attribute name in argparser for torch device option ( #5433)( #5467).Fixed a bug in multi-agent cooperative training where agents might not receive all of the states of.Fixed the bug where curriculum learning would crash because of the incorrect run_options parsing.Fixed a bug where the critics were not being normalized during training.Fixed a bug where ml-agents code wouldn't compile on platforms that didn't support analytics (PS4/5, XBoxOne) ( #5628).This change should enable ml-agents usage on mac M1 ( #5283, #5519) Update gRPC native lib to universal for arm64 and x86_64.Update Colab notebooks for GridWorld example with DQN illustrating the use of the Python API and how to export to ONNX ( #5643)īug Fixes -agents / (C#).Added minimal analytics collection to LL-API ( #5511).Support inference with deterministic action selection in editor ( #5599).Extra tensors are now serialized to support deterministic action selection in onnx. ![]() The same thing canīe achieved by adding deterministic: true under network_settings of the run options configuration.( #5597) Added a new -deterministic cli flag to deterministically select the most probable actions in policy.Deterministic action selection is now supported during training and inference( #5619).env_params.restarts_rate_limit_period_s (-restarts-rate-limit-period-s).env_params.restarts_rate_limit_n (-restarts-rate-limit-n).env_params.max_lifetime_restarts (-max-lifetime-restarts).Rate & lifetime limits on this are configurable via 3 new yaml options.Changed default behavior to restart crashed Unity environments rather than exiting.Added support for having beta, epsilon, and learning rate on separate schedules (affects only PPO and POCA).Set gym version in gym-unity to gym release 0.20.0 ( #5540).Ml-agents / ml-agents-envs / gym-unity (Python) Added the ability to get a read-only view of the stacked observations ( #5523).Added the capacity to initialize behaviors from any checkpoint and not just the latest one ( #5525).Ml-agents / ml-agents-envs / gym-unity (Python) Minor Changes -agents / (C#) Update Barracuda to 2.3.1-preview ( #5591).Added a new feature to replicate training areas dynamically during runtime.The minimum supported Unity version was updated to 2020.3.Renaming the path that shadows torch with "mlagents/trainers/torch_entities" and update respective imports (#)īug Fixes -agents / (C#) ml-agents / ml-agents-envs.Added suppport for enabling/disabling camera object attached to camera sensor in order to improve performance.Current alternating order is still the default but will be deprecated.Added switch to RayPerceptionSensor to allow rays to be ordered left to right.Versioned under ml-agents-envs package in the future (#) We moved UnityToGymWrapper and PettingZoo API to ml-agents-envs package.Add shared critic configurability for PPO.The minimum supported version of PyTorch was changed to 1.8.0.The minimum supported Python version for ML-agents has changed to 3.8.13.Refactored core modules to make ML-Agents internal classes more generalizable to various RL algorithms.ML-Agents Custom Trainers plugin is an extensible plugin system to define new trainers based on the.Add your trainers to the package using Ml-Agents Custom Trainers plugin.The minimum supported Unity version was updated to 2021.3.Release Notes Major Changes -agents / (C#) NOTE: It is strongly recommended that you use packages from the same release together for the best experience.
0 Comments
Leave a Reply. |