gym-zelda-1

An OpenAI Gym environment for The Legend of Zelda (i.e., Zelda 1) on The Nintendo Entertainment System (NES) based on the nes-py emulator.

Installation

The preferred installation of gym-zelda-1 is from pip:

pip install gym-zelda-1

Usage

Python

You must import gym_zelda_1 before trying to make an environment. This is because gym environments are registered at runtime. By default, gym_zelda_1 environments use the full NES action space of 256 discrete actions. To contstrain this, gym_zelda_1.actions provides three actions lists (RIGHT_ONLY, SIMPLE_MOVEMENT, and COMPLEX_MOVEMENT) for the nes_py.wrappers.BinarySpaceToDiscreteSpaceEnv wrapper. See gym_zelda_1/actions.py for a breakdown of the legal actions in each of these three lists.

from nes_py.wrappers import BinarySpaceToDiscreteSpaceEnv
import gym_zelda_1
from gym_zelda_1.actions import SIMPLE_MOVEMENT
env = gym_zelda_1.make('Zelda1-v0')
env = BinarySpaceToDiscreteSpaceEnv(env, SIMPLE_MOVEMENT)

done = True
for step in range(5000):
    if done:
        state = env.reset()
    state, reward, done, info = env.step(env.action_space.sample())
    env.render()

env.close()

NOTE: gym_zelda_1.make is just an alias to gym.make for convenience.

NOTE: remove calls to render in training code for a nontrivial speedup.

Command Line

gym_zelda_1 features a command line interface for playing environments using either the keyboard, or uniform random movement.

gym_zelda_1 -m <`human` or `random`>

Step

Info about the rewards and info returned by the step method.

Reward Function

TODO

info dictionary

The info dictionary returned by the step method contains the following keys:

Citation

Please cite gym-zelda-1 if you use it in your research.

@misc{gym-zelda-1,
  author = {Christian Kauten},
  title = {{The Legend of Zelda} for {OpenAI Gym}},
  year = {2019},
  publisher = {GitHub},
  howpublished = {\url{https://github.com/Kautenja/gym-zelda-1}},
}