PyTorch direct export
The following is intended for informational purposes only, and may not be incorporated into any contract. No purchasing decisions should be made based on the following materials. Unity is not committing to deliver any functionality, features or code. The development, timing and release of all products, functionality and features are at the sole discretion of Unity, and are subject to change.
If you have accepted to use functional cookies and logged in using your Unity ID at the top right of the page, then sharing feedback is as simple as clicking a card below, selecting a topic’s importance, adding your point of view, and submitting. If you prefer not to accept functional cookies or log in, you will be prompted to enter an email address and validate it, so we know how to reach out when the topic evolves. For more information read the Feedback and privacy terms.
What? ExecuTorch is a new PyTorch library that makes it easier to support Torch models locally. We will integrate with their “send to” functionality, which enables a direct export from Python-based Torch models to the .sentis intermediary representation. This enables direct model import into Unity and removes the need to convert to ONNX.
Why? Today, Unity users must convert Python-based Torch models to ONNX and then import them into Sentis. This adds an additional complex step to the import process, and the ONNX conversion can have compatibility issues. By staying close to the source Torch format, we can open up far more Sentis-compatible AI models from the PyTorch community.
*Note: Don’t worry: we will not remove support for ONNX import.