Rousslan F.J. Dossa

Greetings. Sorry if this is necro bumping, but here are the steps that seem to have helped me. Although my error was tied to the horovod.tensorflow module, it might be...

Greetings. Maybe a little bit too late, but the following steps happened to solve the issue with gotop's CPU lines not showing up correctly, as described in the first messages....

@vwxyzjn Thanks for the great addition. Tried to follow up the instructions to get it to work, but there were a few snags along the way: 1. The poetry rule...

Hello there. You may want to try to uncomment the termination judgement conditions in the "gym_torcs.py" file from line 144 to 160.

Is there any benchmark run for this ? How does it perform ? Anyways, great job @timoklein

@vwxyzjn Hello there. Attempted to test the workflow from scratch by installing poetry 1.2.1 as per their docs using this command: `curl -sSL https://install.python-poetry.org | python3 -`, but running `poetry...

Thanks a lot. This seems to have done the trick now.

> In the CQL trainer, the policy_loss is formulated before the QF_Loss is, but the QF_Loss backprops the policy network before policy_loss does, which causes a Torch error. I don't...

I am afraid that change will break the learning of the policy itself, because the `q_new_actions.detach()` in ```python policy_loss = (alpha*log_pi - q_new_actions.detach()).mean() ``` will also block the gradient flow...

According to the screenshots you have provided, those two are not related. The first one (function definiton) uses the low-level API of Tensorflow to create the network, while the second...