Skip to main content

Super Mario Bros. for OpenAI Gym

Project description

gym-zelda-1

BuildStatus PackageVersion PythonVersion Stable Format License

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 constrain this, gym_zelda_1.actions provides an action list called MOVEMENT (20 discrete actions) for the nes_py.wrappers.JoypadSpace wrapper.

from nes_py.wrappers import JoypadSpace
import gym_zelda_1
from gym_zelda_1.actions import MOVEMENT

env = gym_zelda_1.make('Zelda1-v0')
env = JoypadSpace(env, 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: The reward function is a complicated work in progress.

info dictionary

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

Key Type Description
current_level int The current level Link is in (0 for overworld)
x_pos int Link's x position in the screen (from the left)
y_pos int Link's y position in the screen (from the top)
direction str Link's direction as one of {"N", "S", "E", "W"}
has_candled bool Whether Link has candled the current room
pulse_1 str The signal playing through pulse 1
pulse_2 str The signal playing through pulse 2
killed_enemies int The number of enemies killed
number_of_deaths int The number of times Link has died
sword str The kind of sword Link has
number_of_bombs int The number of bombs in Link's inventory
arrows_type str The kind of arrows in Link's inventory
has_bow bool Whether Link has the bow in his inventory
candle_type str The type of candle in Link's inventory
has_whistle bool Whether Link has the whistle in his inventory
has_food bool Whether Link has food in his inventory
potion_type str The type of potion in Link's inventory
has_magic_rod bool Whether Link has the magic rod in his inventory
has_raft bool Whether Link has the raft in his inventory
has_magic_book bool Whether Link has the magic book in his inventory
ring_type str The type of ring in Link's inventory
has_step_ladder bool Whether Link has the step ladder in his inventory
has_magic_key bool Whether Link has the magic key in his inventory
has_power_bracelet bool Whether Link has the power bracelet in his inventory
has_letter bool Whether Link has the letter in his inventory
is_clock_possessed bool Whether the clock is possessed
rupees int The number of rupess Link has collected
keys int The number of keys in Link's inventory
heart_containers int The number of heart containers that Link has
hearts float The number of remaining health Link has
has_boomerang bool Whether Link has the boomerang in his inventory
has_magic_boomerang bool Whether Link has the magic boomerang in his inventory
has_magic_shield bool Whether Link has the magic shield in his inventory
max_number_of_bombs int The maximum number of bombs Link can carry

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}},
}

References

The following references contributed to the construction of this project.

  1. The Legend of Zelda: RAM Map. Data Crystal ROM Hacking.
  2. The Legend of Zelda: Memory Addresses. NES Hacker.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

gym_zelda_1-0.2.2.tar.gz (77.7 kB view details)

Uploaded Source

Built Distribution

gym_zelda_1-0.2.2-py2.py3-none-any.whl (75.1 kB view details)

Uploaded Python 2 Python 3

File details

Details for the file gym_zelda_1-0.2.2.tar.gz.

File metadata

  • Download URL: gym_zelda_1-0.2.2.tar.gz
  • Upload date:
  • Size: 77.7 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/1.13.0 pkginfo/1.5.0.1 requests/2.21.0 setuptools/41.0.1 requests-toolbelt/0.9.1 tqdm/4.28.1 CPython/3.7.3

File hashes

Hashes for gym_zelda_1-0.2.2.tar.gz
Algorithm Hash digest
SHA256 48d9a736072c60e2ca123a7f71b5f7de1b2fcb301a5b229e61b0d5c115b29fee
MD5 9af41d26b1b45e9ce9490e3258c6edbf
BLAKE2b-256 d9d624393b735a5627c478292ee0db5811bf673cd1674683dc20c95c68cd907b

See more details on using hashes here.

File details

Details for the file gym_zelda_1-0.2.2-py2.py3-none-any.whl.

File metadata

  • Download URL: gym_zelda_1-0.2.2-py2.py3-none-any.whl
  • Upload date:
  • Size: 75.1 kB
  • Tags: Python 2, Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/1.13.0 pkginfo/1.5.0.1 requests/2.21.0 setuptools/41.0.1 requests-toolbelt/0.9.1 tqdm/4.28.1 CPython/3.7.3

File hashes

Hashes for gym_zelda_1-0.2.2-py2.py3-none-any.whl
Algorithm Hash digest
SHA256 fa13831b57f2bfafdeddd032d68ff2a35989256dd32af5f7edf2768f1a0db19a
MD5 5ad4abee92d3aa15c2375e47f9cc8994
BLAKE2b-256 169afa3c256620efcb1cde9992471d47cf903b7fdaf5bb962ac5918bcf31a210

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page