Hugging Face Targets Real-World PyTorch Robotics Projects with the Open Source LeRobot
"The goal," the AI company says, "is to lower the barrier to entry to robotics."
Artificial intelligence (AI) specialist Hugging Face has launched a new project to offer state-of-the-art resources for those working with PyTorch-powered real-world robotics: LeRobot.
"LeRobot aims to provide models, datasets, and tools for real-world robotics in PyTorch," the company explains of its latest project launch. "The goal is to lower the barrier for entry to robotics so that everyone can contribute and benefit from sharing datasets and pre-trained models. LeRobot contains state-of-the-art approaches that have been shown to transfer to the real-world with a focus on imitation learning and reinforcement learning."
At its launch, LeRobot includes a set of pre-trained models, datasets with human-collected demonstrations, and simulated environments — but the company has promised that support for real-world robotics will follow "in the coming weeks," with the ability to target "the most affordable and capable robots out there."
Hugging Face has also released a set of examples demonstrating how LeRobot can be used, including one that automatically downloads data from Hugging Face and visualizes episodes from the selected dataset and another which demonstrates how to use LeRobot's training script to train any policy.
LeRobot is now available on Hugging Face and GitHub, under the permissive Apache 2.0 license.