/deep-RL-course-labs

This is my practice for Deep RL course (https://huggingface.co/learn/deep-rl-course/unit0/introduction by Hugging Face

Primary LanguageJupyter Notebook

deep-RL-course-labs

Unit 1 repository: k0T0z/ppo-LunarLander-v2

Bonus unit 1 repository: k0T0z/ppo-Huggy

Unit 2 repositories: k0T0z/q-FrozenLake-v1-4x4-noSlippery, k0T0z/q-Taxi-v3-500

Unit 3 repository: k0T0z/dqn-SpaceInvadersNoFrameskip-v4

Unit 3 building:

  1. Clone the repository.
git clone https://github.com/k0T0z/deep-RL-course-labs.git
  1. Change directory.
cd deep-RL-course-labs
  1. Download the requirements.
pip install -r unit3-requirements.txt
  1. Run the training.
python -m rl_zoo3.train --algo dqn --env SpaceInvadersNoFrameskip-v4  -f logs/  -c dqn.yml
  1. Run the evaluation for 5000 steps.
python -m rl_zoo3.enjoy  --algo dqn  --env SpaceInvadersNoFrameskip-v4  --no-render  --n-timesteps 5000  --folder logs/
  1. Upload the model to the Hugging Face Hub. Replace _____ with your username.
python -m rl_zoo3.push_to_hub  --algo dqn  --env SpaceInvadersNoFrameskip-v4  --repo-name dqn-SpaceInvadersNoFrameskip-v4 -orga _____ -f logs/

Unit 4 building:

  1. Clone the repository.
git clone https://github.com/k0T0z/deep-RL-course-labs.git
  1. Change directory.
cd deep-RL-course-labs
  1. Download the requirements.
pip install -r unit4-requirements.txt
  1. Run the notebook.