Skip to content

Implementacion of NEAT algorithm, PPO algorithm and Genetic algorithm to train agents in the game Airstriker-Genesis, using gym-retro environment.

Notifications You must be signed in to change notification settings

GabrielCande/AirstrikerGenesisAI

Repository files navigation

- Airstriker-Genesis AI -

Repository with implemented NEAT, PPO and Genetic agents for the game Airstriker-Genesis

Dependencies

To run the code you must have the following libraries installed:

gym-retro == 0.25.2

neat-python == 0.92

numpy >= 1.24.4

pygame == 2.5.2

shimmy == 0.2.1

stable_baselines3 == 2.3.2

Note:

Python supported versions: 3.6, 3.7, 3.8

Recommended: Python 3.8.10 or Python 3.7.2

The PPO agents generated are in the google drive link, inside ppo_agents folder, and, just for diclaimer, the ppo agents need to be loaded zipped into the code, do not unzip their folders

Video:

Generated Agents Playing - Video

About

Implementacion of NEAT algorithm, PPO algorithm and Genetic algorithm to train agents in the game Airstriker-Genesis, using gym-retro environment.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages