sharing pretrained PPO and DQN models

[kevinboy01]

Hi, thanks for sharing Pushworld!
I have trained PPO and DQN on level0 using stable baselines3 python package during the night on cpu windows pc for 10 hours. The agents still struggle to solve the level0 puzzles during testing. I am wondering if you can share the pretrained PPO and DQN models trained on level0.