PPO playing QbertNoFrameskip-v4 from https://github.com/sgoodfriend/rl-algo-impls/tree/2067e21d62fff5db60168687e7d9e89019a8bfc0
5c87b65
| ALGO="ppo" | |
| ENVS="CartPole-v1 MountainCar-v0 MountainCarContinuous-v0 Acrobot-v1 LunarLander-v2" | |
| BENCHMARK_MAX_PROCS="${BENCHMARK_MAX_PROCS:-3}" | |
| bash scripts/train_loop.sh -a $ALGO -e "$ENVS" | xargs -I CMD -P $BENCHMARK_MAX_PROCS bash -c CMD | |