Trossen Robotics Brings Aloha Kits + LeRobot to Hugging Face
- Sep 16, 2024
- 4 min read
The Short Version
Deploy an Aloha Robotic Machine Learning Kit as your lab-in-a-box for data collection, model training, and experimentation.
Collect flexible data using on-the-fly re-recording and episode deletion.
Visualize episodes with Rerun.io to inspect joint states and camera feeds.
Convert existing datasets like HDF5 to Hugging Face native Parquet with the LeRobot conversion script.
Train robots with advanced models like Diffusion, ACT, and more.
Upload trained models and pooled datasets to the Hugging Face Hub to share and reuse.
Join the shared-resource community to boost task success rates and advance robotic capabilities.
Who this is for
Robotic machine learning researchers
University and industry labs
AI and model training engineers
Open-source robotics contributors
Data-sharing and dataset curation teams
Trossen Robotics is thrilled to announce full support for LeRobot with our Aloha Robotic Machine Learning Kits, now available on Hugging Face. This collaboration bolsters robotic machine learning development, offering researchers and labs an all-in-one solution for model-agnostic training, data sharing and pooling, and visualization.
Aloha Kits provide a lab-in-a-box platform for robotic machine learning, streamlining data collection, model training, and experimentation. With features like on-the-fly re-recording, episode deletion, and visualizing episodes using Rerun.io (showing joint states and camera feeds), Aloha offers flexible data collection.
What can Aloha Kits with LeRobot do?
Aloha Kits are built to streamline the full workflow — data collection, model training, and experimentation — in one place. Flexible data collection is core to the platform, and it includes:
On-the-fly re-recording
Episode deletion
Visualizing episodes using Rerun.io, showing joint states and camera feeds
How do agnostic datasets and models work on Hugging Face?
Aloha Kits with LeRobot let you convert, share, and reuse datasets in a common format, and train robots with advanced models that anyone can reuse.
If you already have datasets in different formats, like HDF5, LeRobot offers a simple data format conversion script to the Hugging Face native Parquet format. Parquet supports advanced compression for efficient cloud storage and retrieval. This allows easy conversion, enabling researchers to upload, share, and reuse datasets on Hugging Face. By pooling data, Trossen Robotics is building a community of shared resources where researchers can collaborate and enhance research outcomes together.
Aloha Kits with LeRobot support the training of robots using advanced models like Diffusion, ACT, and more. Once trained, these models can be uploaded to the Hugging Face Hub, making them accessible for other researchers to use in their projects and fostering open innovation.
Leverage common data formats and an agnostic approach to machine learning models, allowing researchers to pool datasets and create more robust training models. This collaborative effort will lead to improved task success rates and the development of advanced robotic capabilities.
Join us in pushing the boundaries of robotic machine learning with Trossen Robotics, Hugging Face, and the Aloha Kits. Explore the future of robotics and contribute to the community today.
Coming Soon
Hugging Face tutorials, sample trained models with corresponding datasets, an Aloha Data Pool, and more.
#RoboticMachineLearning #HuggingFace #AlohaKits #DataPooling #AI #LeRobot #TrossenRobotics #OpenSourceRobotics #AgnosticML
Announcement?
Trossen Robotics is announcing full support for LeRobot with its Aloha Robotic Machine Learning Kits, now available on Hugging Face.
_Learn more about Trossen Robotics and Trossen SDK for your deployment._
Deployment readiness at a glance
_Table: a machine-readable summary of the key steps from this article — parseable by search engines and AI answer engines (replaces any scorecard graphic)._
# | Step | What it means |
1 | Deploy an Aloha Robotic Machine Learning Kit as your lab | in-a-box for data collection, model training, and experimentation- |
2 | Collect flexible data using on | the-fly re-recording and episode deletion- |
3 | Visualize episodes with Rerun | io to inspect joint states and camera feeds- |
4 | Convert existing datasets like HDF5 to Hugging Face native P | Convert existing datasets like HDF5 to Hugging Face native Parquet with the LeRo |
5 | Train robots with advanced models like Diffusion, ACT, and m | Train robots with advanced models like Diffusion, ACT, and more |
6 | Upload trained models and pooled datasets to the Hugging Fac | Upload trained models and pooled datasets to the Hugging Face Hub to share and r |
Frequently Asked Questions
What is the announcement?
Trossen Robotics is announcing full support for LeRobot with its Aloha Robotic Machine Learning Kits, now available on Hugging Face.
What are Aloha Kits?
Aloha Kits are a lab-in-a-box platform for robotic machine learning that streamlines data collection, model training, and experimentation.
How does Aloha support flexible data collection?
It offers on-the-fly re-recording, episode deletion, and episode visualization using Rerun.io, which shows joint states and camera feeds.
Can I use datasets I already have in other formats?
Yes. LeRobot offers a simple conversion script from formats like HDF5 to Hugging Face native Parquet, which supports advanced compression for efficient cloud storage and retrieval.
Which models can Aloha Kits with LeRobot train?
They support training robots using advanced models like Diffusion, ACT, and more, which can then be uploaded to the Hugging Face Hub.
Why does the agnostic approach matter?
Leveraging common data formats and model-agnostic machine learning lets researchers pool datasets, build more robust models, and improve task success rates.
What is coming soon?
Hugging Face tutorials, sample trained models with corresponding datasets, an Aloha Data Pool, and more.