[0.4.6] - 2024-01-30

Added

• Added ema_ratio parameter to rockrl.tensorflow.ppo to control the exponential moving average ratio of the old and new networks

[0.4.5] - 2024-01-17

Changed

• Changed rockrl.tensorflow.ppo act function to use training mode while training and evaluation mode while evaluating

Added

• Added close() method to rockrl.tensorflow.ppo to close tensorBoardLogger after training

[0.4.3] - 2024-01-05

Changed

• Updated rockrl.tensorflow.ppo to use threading while logging, to save time

[0.4.2] - 2024-01-02

Changed

• Changed rockrl.utils.VectorizedEnv to return info dictionaries in list instead of tuples

[0.4.1] - 2023-12-08

Changed

• Removed tensorflow_probability dependency

[0.0.4] - 2022-10-10

Added

• Initial RockRL package release
• Tensorflow PPO implementation for discrete and continuous action spaces
• Added Memory (one environment) and MemoryHandler (multiple environments) object to rockrl.utils package
• Examples for LunarLander-v2, BipedalWalker-v3 and BipedalWalkerHardcore-v3 environments