Antonin RAFFIN

> But the question is: how important are those options? I presume tests aren't important for the end user (I can be wrong, of course), The only requirement is `stable-baselines3`,...

should be fixed with the rl zoo released as a package (for minimal support, you don't need `requirements.txt`, `pip install rl_zoo3` or `pip install -e .` from source)

Hello, I'm unsure about such feature. On one side, it seems to be a reasonable (even though unconventional) request. On the other side there are some behaviors that may be...

Hello, I've got mixed feeling about this. On one hand, running several times the same set of hyperparams is the correct way to go, on the other hand, as we...

> To clarify, this is taking the 10 best hyperparameters and running them a bunch, not running each 1 time in training. That's much more computationally reasonable. Oh, I see,...

Hello, Please take a look at SAC hyperparameters, it is using a time feature wrapper (hence 29).

>I did a search and couldn't find 'time feature wrapper'. I wrote code to remove the 29th feature. Is this wrapper appended? Is there an easier or more correct solution?...

Hello, thanks for suggesting the feature to Optuna. If I understand, you want to automate something that is similar to https://github.com/optuna/optuna-examples/blob/main/rl/sb3_simple.py#L44 ? My main concern with the hack is the...

Good point. Overall, it should not make a big difference as the main point is to normalize the reward magnitude. But for consistency, I agree that gamma should be updated....

>only hyperparmeters that are searched by default can have their params dict be copied and reused due to naming issues well also some params that are searched cannot be copied...