maruya24/pytorch_robotics_transformer

A PyTorch re-implementation of the RT-1 (Robotics Transformer)

42
/ 100
Emerging

This project helps robotics engineers and researchers quickly integrate and experiment with the Robotics Transformer (RT-1) model using familiar PyTorch and OpenAI Gym environments. It takes in visual observations from a robot's camera and other sensor data, and outputs precise action commands to control the robot's movements and tasks. This is for professionals working on real-world robot control and learning from demonstrations.

No commits in the last 6 months.

Use this if you are a robotics engineer or researcher building robot control systems and prefer to work within the PyTorch and OpenAI Gym ecosystem, avoiding TensorFlow dependencies.

Not ideal if you need the full training and inference utilities that rely on the TF-Agents library, as this PyTorch re-implementation focuses on core model compatibility.

robotics-control robot-learning autonomous-systems reinforcement-learning robot-manipulation
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 8 / 25
Maturity 16 / 25
Community 18 / 25

How are scores calculated?

Stars

51

Forks

15

Language

Python

License

Apache-2.0

Last pushed

Oct 18, 2023

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/maruya24/pytorch_robotics_transformer"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.