Viktor Makoviichuk

> @ViktorM @Denys88 the example above assumes that you are using a frozen network. You can't optimise the weights of this network because `rl_games` has `torch.infernce()` context for doing running...

Thanks @gonzalesMK for the finding! Could you please create a PR with the fix? Also hopefully their API is much more stable now I can take a look again into...

Hi @Mayankm96, Could you remind what gymnasium version is needed? If I am not mistaken there were some conflicting changes in the latest gymnasium and some libraries decided not to...

Hi Kelly. We'll add 3.11 and 3.12 in the next release. rl-games are fully compatible with these versions, I am using 3.12 with a branch that is going to be...

Unfortunately I found that while rl_games already works with 3.11 and 3.12, a full switch to 3.11 breaks some older compatibilities. Having a discussion with @Denys88 if we should fully...

@kellyguo11 sorry for the delay. A number of minor issues were found during testing after switching to Gymnasioum for Python >=3.9 while kipping Gym for Python 3.8 for Isaac Gym...

Thank you for the report! It was fixed in https://github.com/Denys88/rl_games/tree/VM/vision branch, and will be merged to master shortly and be a part of the new release soon.