trained RL agent

[wardstoneX]

Hi, anyone knows any trained RL agent(open source ?) that is trained in Carla and drives good(idk how good is defined) ? i wanted to plug this into scenarios and measure stuff if possible.