mmurooka / eipl

EIPL: Embodied Intelligence with Deep Predictive Learning

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

What's EIPL?

EIPL (Embodied Intelligence with Deep Predictive Learning) is a library for robot motion generation using deep predictive learning developed at Ogata Laboratory, Waseda University. Highlighted features include:

  • Full documentation for the systematic understanding of deep predictive learning
  • Easy model training: Includes sample datasets, source code, and pre-trained weights
  • Applicable to real robots: Generalized motion can be acquired with small data sets

Install

pip install -r requirements.txt
pip install -e .

Citation

@article{suzuki2023deep,
  author    = {Kanata Suzuki and Hiroshi Ito and Tatsuro Yamada and Kei Kase and Tetsuya Ogata},
  title     = {Deep Predictive Learning : Motion Learning Concept inspired by Cognitive Robotics}, 
  booktitle = {arXiv preprint arXiv:2306.14714},
  year      = {2023},
}

LICENSE

AGPL-3.0

About

EIPL: Embodied Intelligence with Deep Predictive Learning

License:GNU Affero General Public License v3.0


Languages

Language:Python 97.1%Language:CMake 2.7%Language:Shell 0.2%