Orion is an open-source, community-driven framework dedicated to Provable Machine Learning. It provides essential components and a new ONNX runtime for building verifiable Machine Learning models using STARKs.
ONNX (Open Neural Network Exchange), is an open-source standard created to represent deep learning models. The aim of its development was to enable interoperability among diverse deep learning frameworks, like TensorFlow or PyTorch. By offering a universal file format, ONNX allows models trained in one framework to be readily applied in another for inference, eliminating the need for model conversion.
Ensuring compatibility with ONNX operators facilitates integration into the ONNX ecosystem. This enables researchers and developers to pre-train models using their preferred framework, before executing verifiable inferences with Orion.
You can check our official docs here.
- 🧱 Framework: The building blocks for Verifiable Machine Learning models.
- 🏛 Hub: A curated collection of ML models and spaces built by the community using Orion framework.
- 🎓 Academy: Resources and tutorials for learning how to build ValidityML models using Orion.
For a detailed list of changes, please refer to the CHANGELOG file.
Join the community and help build a safer and transparent AI in our Discord!
For a full list of all authors and contributors, see the contributors page.
This project is licensed under the MIT license.
See LICENSE for more information.
Thanks goes to these wonderful people (emoji key):
This project follows the all-contributors specification. Contributions of any kind welcome!