stable-baselines3-contrib

stable-baselines3-contrib copied to clipboard

a bug with MultiDiscrete action space when actions are not same size

23

when the environment has a action space that each action has different size, like this: `self.action_space = MultiDiscrete([3,2])` and the action masker is like this for example: `a = [[True,...

vahidqo

[Bug] An error in MaskPPO training

19

System Info Describe the characteristic of your environment: Describe how the library was installed: pip sb3-contrib=='1.5.1a9' Python: 3.8.13 Stable-Baselines3: 1.5.1a9 PyTorch: 1.11.0+cu102 GPU Enabled: False Numpy: 1.22.3 Gym: 0.21.0 My...

Yangxiaojun1230

## Description ## Context - [ ] I have raised an issue to propose this change ([required](https://github.com/Stable-Baselines-Team/stable-baselines3-contrib/blob/master/CONTRIBUTING.md)) ## Types of changes - [ ] Bug fix (non-breaking change which fixes...

araffin

Implement PPG

10

As discussed in [sb3#346](https://github.com/DLR-RM/stable-baselines3/issues/346), I'd like to merge [an existing implementation of the PPG algorithm](https://github.com/janEbert/sb3-ppg). I'm unsure about the two SDE-related calls [here](https://github.com/janEbert/sb3-ppg/blob/14a0e680d6a3b6ccb955e62037fe2c7bf8693c2e/ppg/ppg.py#L239-L240) and [here](https://github.com/janEbert/sb3-ppg/blob/14a0e680d6a3b6ccb955e62037fe2c7bf8693c2e/ppg/ppg.py#L257-L258); I just oriented myself on...

janEbert

## Description ## Context - [ ] I have raised an issue to propose this change ([required](https://github.com/Stable-Baselines-Team/stable-baselines3-contrib/blob/master/CONTRIBUTING.md)) ## Types of changes - [ ] Bug fix (non-breaking change which fixes...

araffin

[BUG] action masking does not work with VecEnv and MultiDiscrete action space

3

**Describe the bug** I am aware of https://github.com/Stable-Baselines-Team/stable-baselines3-contrib/issues/49#issuecomment-957629253 - but it still does not work. I have investigated the code and this is what I found: When having more than...

clotodex

Change from gamma=0.4 to default in Example Docu

2

Dear sb3-contrib creators, thanks for this awesome repo. I have just one small suggestion for the Documentation which might have a big impact for the user: https://github.com/Stable-Baselines-Team/stable-baselines3-contrib/blob/master/docs/modules/ppo_mask.rst In this Example...

LutzFassl

Implement PPO MPI (SB2 PPO1)

2

MPI can be quite useful to use multiprocessing [full potential](https://twitter.com/hardmaru/status/1260852988475658242) but it is dependency that can be tricky to install.

araffin

[Feature Request] Better support for action masking for vectorized environments

2

**Motivation** Stable-baselines3 (SB3) has introduced support for action masking (see [here](https://sb3-contrib.readthedocs.io/en/master/modules/ppo_mask.html)), which is a great feature. However, this API requires the user to provide an `ActionMasker` wrapper. The issue is...

BolunDai0216

[feature request] Parameterized action spaces.

2

Add a new action space for actions with parameters. [This](https://paperswithcode.com/paper/reinforcement-learning-with-parameterized) paper has an example of parameterized action spaces. The paper also demonstrates the "Q-PAMDP" algorithm with parameterized action spaces.

Cheeseboy8020