Skip to content

sled-group/Teachable_RL

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

4 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

🤖 Teaching Embodied Reinforcement Learning Agents: Informativeness and Diversity of Language Use

Paper

Overview

⚒️ Environment Setup

HomeGrid

Create conda env.

conda create -n homegrid-ltdt python=3.8
conda activate homegrid-ltdt

Install packages.

cd <path/to/homegrid>
pip install homegrid
pip install -r requirements.txt --no-deps

ALFWorld

Create conda env.

conda create -n alfworld-ltdt python=3.9
conda activate alfworld-ltdt

Install packages.

cd <path/to/alfworld>
pip install -e .[alfworld]
pip install -r requirements.txt --no-deps

Messenger

Create conda env.

conda create -n messenger python=3.85
conda activate messenger

Install packages

git clone [email protected]:ahjwang/messenger-emma.git
cd <path/to/messenger/messenger-emma>
pip install -e . 
pip install git+https://github.com/ahjwang/py-vgdl
cd <path/to/messenger>
pip install -r requirements.txt --no-deps

Trouble Shoot: If you have difficulty installing packages from messenger-emma for pygame, comment the requirement for pygame version in messenger-emma/setup.py.

MetaWorld

Create conda env.

conda create -n messenger python=3.85
conda activate messenger
cd <path/to/metaworld>
pip install -r requirements.txt --no-deps

🔧 Model Training & Eval

For training and evaluation, please refer to the README in each folder.

🤗 Data & Model ckpts

Data and model checkpoints are available here: Data & Model Checkpoints

Acknowledgement

This work was supported by NSF IIS-1949634 and has benefited from the Microsoft Accelerate Foundation Models Research (AFMR) grant program. We would like to thank the anonymous reviewers for their valuable comments and suggestions.

Citation

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages