Sentis overview | Sentis | 2.6.1

3 min read Original article ↗
Note

In Sentis 2.4, the package display name changed from Inference Engine to Sentis. You can find the Sentis package by searching Sentis or Inference Engine in the Package Manager. You don't need to change any code in your project as this is a change to the package display name only.

Sentis is a neural network inference library for Unity. It lets you import trained neural network models into Unity and run them in real-time with your target device’s compute resources, such as central processing unit (CPU) or graphics processing unit (GPU).

Sentis supports real-time applications across all Unity-supported platforms.

The package is officially released and available to all Unity users through the Package Manager.

Tip

Prior experience with machine learning frameworks like TensorFlow or PyTorch is helpful, but not required. It can make it easier to understand how to work with models in Sentis.

Section Description
Get started Learn how to install Sentis, explore sample projects, and understand the Sentis workflow.
Create a model Create a runtime model by importing an ONNX model file or using the Sentis model API.
Run a model Create input data for a model, create an engine to run the model, and get output.
Use Tensors Learn how to get, set, and modify input and output data.
Profile a model Use Unity tools to profile the speed and performance of a model.

Supported platforms

Sentis supports all Unity runtime platforms.

Performance might vary based on:

  • Model operators and complexity

  • Hardware and software platform constraints of your device

  • Type of engine used

    For more information, refer to Models and Create an engine.

Supported model types

Sentis supports most models in Open Neural Network Exchange (ONNX) format with an opset version between 7 and 25. For more information, refer to Supported models and Supported ONNX operators.

Sentis supports most models in LiteRT (formerly TensorFlow Lite) format. For more information, refer to Supported LiteRT operators.

Sentis supports most models (exported programs) in PyTorch format if they are decomposed to Core ATen IR operators. For more information, refer to Supported PyTorch operators.

Places to find pre-trained models

There are various sources to find pre-trained models, which might either be available in the ONNX format, LiteRT format, or in a format that you can convert. Examples include:

If you want to train your own models, refer to the following links:

Additional resources