Hi @frederikcodes!
I was able to make some updates on my new fork. Would you have a chance to review this Jupyter Notebook and simulation video?
https://github.com/AdaptLLC/intelevator/blob/main/Elevator_Modell_Simulation/demo.ipynb
https://github.com/AdaptLLC/intelevator/blob/main/Elevator_Modell_Simulation/elevator_sim_comparison.mp4
And there's the model after 200 training episodes:
https://github.com/AdaptLLC/intelevator/blob/main/Elevator_Reinforcement_Training/checkpoints_v2/ppo_elevator_episode_200.zip
I was able to run the RL training to get similar results to what you listed in the original README.md. This was my personal goal to get oriented around the project and how to run it. I'm at a point where I'd like to compare results with you, so stay in sync with your original research direction.
Ultimately, I'd like this to serve as a backend routing engine.
I hope all is well. Looking forward to collaborating here.
Hi @frederikcodes!
I was able to make some updates on my new fork. Would you have a chance to review this Jupyter Notebook and simulation video?
https://github.com/AdaptLLC/intelevator/blob/main/Elevator_Modell_Simulation/demo.ipynb
https://github.com/AdaptLLC/intelevator/blob/main/Elevator_Modell_Simulation/elevator_sim_comparison.mp4
And there's the model after 200 training episodes:
https://github.com/AdaptLLC/intelevator/blob/main/Elevator_Reinforcement_Training/checkpoints_v2/ppo_elevator_episode_200.zip
I was able to run the RL training to get similar results to what you listed in the original README.md. This was my personal goal to get oriented around the project and how to run it. I'm at a point where I'd like to compare results with you, so stay in sync with your original research direction.
Ultimately, I'd like this to serve as a backend routing engine.
I hope all is well. Looking forward to collaborating here.