0.0.6

What's Changed

• Fix numpy typing

Full Changelog: 0.0.5...0.0.6

0.0.5

What's Changed

• Reply buffer accepts Experience data class

Full Changelog: 0.0.4...0.0.5

0.0.4

What's Changed

• Simplify release flow in #56
• Change log level in #58
• Add save_model function in #63
• Refactor collect_one_epoch_experience function in OnPolicyAlgorithm in #65
• Move last reward bootstrap to train function for on-policy in #67
• Refactor collect one epoch experience in off policy algorithm in #68
• Use action selection function in #69
• Delete common folder in #70
• Use same data class for one epoch experience in #72
• Bump up Python version to 3.10 in #77
• Refactor mypy in #78
• Refactor comments in #81
• Bump down python version in #84
• Update example in #85
• Refactor experience calculations in #90
• Refactor number variable names in #91
• Refactor Replybuffer in #92
• Implement samplers in #93
• Implement seed manager in #95
• Extract ReplyBuffer in #97
• Implement evaluator in #99
• Add get_action_tensor and get_action_numpy in #101
• Always log for tensorboard in #103
• Refactor train functions in #104
• Delete base algorithms in #106
• Replace select_action_with_noise with add_noise_to_get_action in #107
• Improve logging of train function in #108
• Add helper functions to train function in #109
• Add evaluation to integration tests in #111
• Add metrics manager in #112
• Improve model saving in #113
• Benchmarking in #116
• Fix ddpg and td3 in #118
• Simplify seed management in #120
• Update gym env versions in #121
• Improve release flow in #122
• Fix release flow in #124

Full Changelog: 0.0.3...0.0.4

0.0.3

What's Changed

• Split policy into base policy, stochastic policy and categorical policy in #25
• Implement DDPG in #26
• Create MLP class in #28
• Refactor comments in #30
• Refactor DDPG in #31
• Store each variables on list in #32
• Implement TD3 in #33
• Fix typing in #35
• Logging improvement in #37
• Support continuous action spaces in #39
• Adopt the src layout in #41
• Improve packaging in #46
• set up GitHub Actions for PyPI and Test PyPI in #47
• Add linters in #50
• Add integration test in #52
• Set up mypy in #53

Full Changelog: 0.0.2...0.0.3

0.0.2

Reinforcement Learning Replications 0.0.2