Nicola Russo

Hi, I think there is an issue with GPU training. I'll test it, thanks for pointing this out.

@darshisen: Unfortunately this feature is not tested nor properly supported. However, according with TF [docs](https://www.tensorflow.org/guide/gpu#overview), you should be able to just change this [line](https://github.com/nicrusso7/rex-gym/blob/82dea26bcd8896da06240bcc3abd4de5b4696430/rex_gym/playground/trainer.py#L109). Any test is appreciated!

Hello, This looks like an issue with the rex-gym PyPi package and its dependencies, may I ask to install the lib from source? Just clone this repository and run (from...

Thanks for pointing this out, I’ll fix the dependencies tree ASAP. In the meanwhile you can just install it with: `pip install --no-dependencies .`

Hi, thanks for reaching me. This project is more an experiment, started from the Minitaur open source examples you can find in the pyBullet repository. Frankly, I don’t have a...

Hi, thanks for raising this issue. This library is not properly tested on Windows, I’ll try to reproduce this issue ASAP. Thanks!

It seems pickling issue in multiprocessing on windows. Unfortunately this library isn't supporting Windows at the moment. I will release a docker image in the near future, but for now...

Hi @iftikarmuhammad, the `wrappers.ExternalProcess` creates a new process for each agent (for lock free parallelism) meanwhile in the else statement, all the agents are created in the same process. This...

I've just pushed a hotfix for this, it should force single process now on Windows (please install from source for now).

Hi, thanks for your contribution. Do you have any evidence on how this improves the performance? Thanks!