minqi / babyai

BabyAI platform. A testbed for training agents to understand and execute language commands.

BabyAI Platform

A platform for simulating language learning with a human in the loop. This is an ongoing research project based at Mila.

Contents:

• Citation
• Replicating ICLR19 Results
• Installation
• Usage
• Codebase Structure
• Levels
• Training and Evaluation
• Contributing
• Troubleshooting
• About

Citation

If you use this platform in your research, please cite:

@inproceedings{
  babyai_iclr19,
  title={Baby{AI}: First Steps Towards Grounded Language Learning With a Human In the Loop},
  author={Maxime Chevalier-Boisvert and Dzmitry Bahdanau and Salem Lahlou and Lucas Willems and Chitwan Saharia and Thien Huu Nguyen and Yoshua Bengio},
  booktitle={International Conference on Learning Representations},
  year={2019},
  url={https://openreview.net/forum?id=rJeXCo0cYX},
}

Replicating ICLR19 Results

The master branch of this repository is updated frequently. If you are looking to replicate or compare against the results from the ICLR19 BabyAI paper, please use the docker image, demonstration dataset and source code from the iclr19 branch of this repository.

Installation

Requirements:

• Python 3.5+
• OpenAI Gym
• NumPy
• PyTorch 0.4.1+
• blosc

Start by manually installing PyTorch. See the PyTorch website for installation instructions specific to your platform.

Then, clone this repository and install the other dependencies with pip3:

git clone https://github.com/mila-iqia/babyai.git
cd babyai
pip3 install --editable .

Installation using Conda (Alternative Method)

If you are using conda, you can create a babyai environment with all the dependencies by running:

git clone https://github.com/mila-iqia/babyai.git
cd babyai
conda env create -f environment.yaml
source activate babyai

After that, execute the following commands to setup the environment.

cd ..
git clone https://github.com/maximecb/gym-minigrid.git
cd gym-minigrid
pip install --editable .

The last command installs the repository in editable mode. Move back to the babyai repository and install that in editable mode as well.

cd ../babyai
pip install --editable .

BabyAI Storage Path

Add this line to .bashrc (Linux), or .bash_profile (Mac).

export BABYAI_STORAGE='/<PATH>/<TO>/<BABYAI>/<REPOSITORY>/<PARENT>'

where /<PATH>/<TO>/<BABYAI>/<REPOSITORY>/<PARENT> is the folder where you typed git clone https://github.com/mila-iqia/babyai.git earlier.

Models, logs and demos will be produced in this directory, in the folders models, logs and demos respectively.

Usage

To run the interactive GUI application that illustrates the platform:

scripts/manual_control.py

The level being run can be selected with the --env option, eg:

scripts/manual_control.py --env BabyAI-UnlockPickup-v0

The Levels

Documentation for the ICLR19 levels can be found in docs/iclr19_levels.md. There are also older levels documented in docs/bonus_levels.md.

Pixel Observations

Please note that the default observation format is a partially observable view of the environment using a compact encoding, with 3 input values per visible grid cell, 7x7x3 values total. These values are not pixels. If you want to obtain an array of RGB pixels as observations instead, use the RGBImgPartialObsWrapper. You can use it as follows:

import babyai
from gym_minigrid.wrappers import *
env = gym.make('BabyAI-GoToRedBall-v0')
env = RGBImgPartialObsWrapper(env)

This wrapper, as well as other wrappers to change the observation format can be found here.

About this Project

BabyAI is an open-ended grounded language acquisition effort at Mila. The current BabyAI platform was designed to study data-effiency of existing methods under the assumption that a human provides all teaching signals (i.e. demonstrations, rewards, etc.). For more information, see the ICLR19 paper.