Abhay Rawat

I can also confirm this. I ran the evaluation script with the checkpoint provided and was getting similar results! 

@Hsgngr I updated the link!!

As of now I have zeroed in to this snippet of code that raises an exception whenever the number of agents is greater than one in the `UnityEnvironment` https://github.com/Unity-Technologies/ml-agents/blob/3c2fa4d8d1cd981e9cef6b2e2fdb2f77757983c3/gym-unity/gym_unity/envs/__init__.py#L267-L272 Is...

Hi @xiaomaogy, So I've currently managed to get the data from multiple agents by bypassing the above mentioned check. For stepping I change the following line https://github.com/Unity-Technologies/ml-agents/blob/3c2fa4d8d1cd981e9cef6b2e2fdb2f77757983c3/gym-unity/gym_unity/envs/__init__.py#L169 using `(-1, action_size)`...

This snippet is actually responsible for sending the observations when some agent reaches its terminal condition. https://github.com/Unity-Technologies/ml-agents/blob/3c2fa4d8d1cd981e9cef6b2e2fdb2f77757983c3/gym-unity/gym_unity/envs/__init__.py#L175-L180 In a multi-agent/vectorized setting would it be okay to return the observations/rewards/dones considering...

@OccupyMars2025 I solved it by installing GCC/++ version `9.4`. You can follow [this](https://linuxize.com/post/how-to-install-gcc-compiler-on-ubuntu-18-04/) link to do the same. Make sure that your `gcc` and `g++` point to the new version...