Quentin Gallouédec

Not sure to understand your question. Next time, please provide a piece of code to help us better understand your question. From the doc, and trying to match your description:...

```python from stable_baselines3 import PPO import gym env = gym.make("Pendulum-v1") ppo = PPO("MlpPolicy", env, device="mps") ppo.learn(total_timesteps=1000) ``` With current pip version of PyTorch (1.12), it raises the following exception: ```log...

> So it is not completely stable for the moment. We may have to wait until the next release... The bug no longer occurs with the new version of PyTorch...

For the moment, consider that SB3 is not compatible with MPS. But we are working on it: #951 Have you seen something in the documentation about float64 and MPS? (You...

Please help us to help you by filling completely the issue template.

Unfortunately, the maintainers don't have time to offer technical support. Check the links [here](https://github.com/DLR-RM/stable-baselines3/blob/master/.github/ISSUE_TEMPLATE/question.md) to find other forums to ask this question. I also suggest that you carefully read the...

Not only the pytest failed, but it caused a Python Fatal Error: ``` (env) quentingallouedec@MacBook-Pro-de-Quentin stable-baselines3 % pytest tests/test_cnn.py =========================================== test session starts ============================================ platform darwin -- Python 3.9.13, pytest-7.1.2,...

Well, I'm pretty sure the problem comes from the fact that the observation is transposed before passing into the CNN of the feature extractor, and this seems to cause some...

What about adding the version warning as a banner on every page? 

Hi, I'm working on this feature, and it's not clear why there is a class dedicated to the case where `n_envs == 1`: `HerReplayBuffer` , and another one for the...