https://github.com/wookayin/toy-montezuma-pycolab

Toy Montezuma's Revenge as Pycolab and Gym environments (working in progress, NOT FULLY USABLE)
https://github.com/wookayin/toy-montezuma-pycolab

Last synced: about 1 month ago
JSON representation

Toy Montezuma's Revenge as Pycolab and Gym environments (working in progress, NOT FULLY USABLE)

• Host: GitHub
• URL: https://github.com/wookayin/toy-montezuma-pycolab
• Owner: wookayin
• License: apache-2.0
• Created: 2018-06-08T23:10:33.000Z (over 6 years ago)
• Default Branch: master
• Last Pushed: 2021-08-02T08:13:24.000Z (over 3 years ago)
• Last Synced: 2023-04-09T09:50:53.679Z (over 1 year ago)
• Language: Python
• Homepage:
• Size: 146 KB
• Stars: 6
• Watchers: 3
• Forks: 2
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
Toy Montezuma's Revenge

=======================

![](assets/ToyMontezumaRevenge.png)

This is a reproduction of the Toy Montezuma's Revenge environment described in

[Deep Abstract Q-Networks (Roderick et al., 2017)](https://arxiv.org/abs/1710.00459).

Rewards:

* +100 for picking up a key

* +300 for opening a door

* +10000 for reaching at the goal location

## Usage

Demo (play by hand):

```bash

python -m mr_pycolab.toy_montezuma

# All rooms are fully observable rather than partially:

python -m mr_pycolab.toy_montezuma --full-observation

```

## OpenAI gym

```python

import mr_pycolab, gym

env = gym.make("ToyMontezumaRevenge-v0")

s = env.reset()   # [11, 11, 5]

actions = ('D', 'U', 'L', 'R', '?')

s, r, done, info = env.step(env.action_space.sample())

```