LiteRT (TensorFlow Lite) Applications
R2026bGenerate code that performs inference using TFLite library and deploy on embedded targets. These examples require the Deep Learning Toolbox™ Interface for LiteRT Library add-on.
Note
The Deep Learning Toolbox Interface for LiteRT Library add-on does not generate code for the deep learning model itself. It generates code that calls the LiteRT interpreter, which executes the model at runtime. This add-on supports C++ code generation only.
Starting in R2026a, to generate C/C++ or CUDA code directly from LiteRT models, use the MATLAB® Coder™ Support Package for PyTorch® and LiteRT Models. See Install MATLAB Coder Support Package for PyTorch and LiteRT Models.
Featured Examples
Generate Code for TensorFlow Lite (TFLite) Model and Deploy on Raspberry Pi
Generate code that uses a TensorFlow™ Lite model for inference.
Deploy Classification Application Using Mobilenet-V3 TensorFlow Lite Model on Host and Raspberry Pi
Generate code for a classification segmentation application that uses TensorFlow Lite model.
Deploy Semantic Segmentation Application Using TensorFlow Lite Model on Host and Raspberry Pi
Generate code for an image segmentation application that uses TensorFlow Lite model.
Deploy Pose Estimation Application Using TensorFlow Lite Model (TFLite) Model on Host and Raspberry Pi
Simulate and generate code for a human pose estimation TFLite model.
Deploy Super Resolution Application That Uses TensorFlow Lite (TFLite) Model on Host and Raspberry Pi
Generate code for a super resolution application that uses a TFLite model for inference.
Audio Event Classification Using TensorFlow Lite on Raspberry Pi
Perform audio event classification on Raspberry Pi® using the YAMNet pretrained deep neural network from the TensorFlow Lite library.
