Skip to content

Use of reinforcement learning with PPO to train an Agent that can beat the first level of Donkey Kong

Notifications You must be signed in to change notification settings

maxogod/AI-Donkey-Kong

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

42 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Donkey Kong AI

video

Sneak peeks

easy ladder

hard ladder

How to use

Install Unity

Unity 6 (6000.0.27f1) Download

Install Conda

MiniConda Download

Install Pytorch and ML-Agents Toolkit

# Create from environment config file
conda env create -f ./conda_env.yaml
conda activate donkey

Alternatively...

conda create -n donkey python=3.10.12 # for mlagents 22 (oct/5/2024)
conda activate donkey

# If using CUDA platform
pip3 install torch~=2.2.1 --index-url https://download.pytorch.org/whl/cu121

# If using AMD ROC platform (only supported in linux)
pip3 install torch~=2.2.1 --index-url https://download.pytorch.org/whl/rocm6.2

# If not using GPU acceleration (CPU only)
pip3 install torch~=2.2.1

python -m pip install mlagents==1.1.0 # If downloading from PyPi, otherwise download it from the ML-Agents gh repo

mlagents-learn --help # verify if it was installed correctly

About

Use of reinforcement learning with PPO to train an Agent that can beat the first level of Donkey Kong

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages