MediaTek Backend

MediaTek backend empowers ExecuTorch to speed up PyTorch models on edge devices that equips with MediaTek Neuron Processing Unit (NPU). This document offers a step-by-step guide to set up the build environment for the MediaTek ExecuTorch libraries.

What you will learn in this tutorial:

• How to export and lower a PyTorch model ahead of time with ExecuTorch for MediaTek devices.

• How to build MediaTek backend and examples.

• How to deploy the exported models on device with ExecuTorch runtime.

Tutorials we recommend you complete before this:

• Introduction to ExecuTorch

• Getting Started

• Building ExecuTorch with CMake

Prerequisites (Hardware and Software)

Host OS

• Linux operating system

Supported Chips:

• MediaTek Dimensity 9300 (D9300)

• MediaTek Dimensity 9400 (D9400)

Software:

• NeuroPilot Express SDK is a lightweight SDK for deploying AI applications on MediaTek SOC devices.

Setting up your developer environment

Follow the steps below to setup your build environment:

1. Setup ExecuTorch Environment: Refer to the Getting Started guide for detailed instructions on setting up the ExecuTorch environment.

2. Setup MediaTek Backend Environment

   pip3 install -r requirements.txt
   

• Install the two .whl downloaded from NeuroPilot Portal

  pip3 install mtk_neuron-8.2.19-py3-none-linux_x86_64.whl
  pip3 install mtk_converter-8.13.0+public-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl
  

• Set evironment variables for building backend

  export NEURON_BUFFER_ALLOCATOR_LIB=<path_to_buffer_allocator>
  

Additionally, make sure to copy NeuronAdapter.h to the following directory: backends/mediatek/runtime/include/api/.

Build

Ahead of time:

Exporting a PyTorch Model for MediaTek Backend:

1. Lower and export the .pte file for on-device execution. The export script samples are povided under example/mediatek/. For example, the following commnad exports the .pte using the scripts provided.

cd executorch

./examples/mediatek/shell_scripts/export_oss.sh mobilenetv3

2. Find the .pte files under the directory named as same as the model.

Runtime:

Build MediaTek Backend for ExecuTorch Runtime

1. Navigate to backends/mediatek/scripts/ directory.

2. Build MediaTek Backend: Once the prerequisites are in place, run the mtk_build.sh script to start the build process:

   ./mtk_build.sh
   

3. MediaTek backend will be built under cmake-android-out/backends/ as libneuron_backend.so.

Build a runner to execute the model on the device:

1. Build the runners and the backend by exedcuting the script:

./mtk_build_examples.sh

2. The runners will be built under cmake-android-out/examples/

Deploying and running on a device

1. Push MediaTek universal SDK and MediaTek backend to the device: push libneuronusdk_adapter.mtk.so and libneuron_backend.so to the phone and export it to the $LD_LIBRARY_PATH environment variable before executing ExecuTorch with MediaTek backend.

   export LD_LIBRARY_PATH=<path_to_usdk>:<path_to_neuron_backend>:$LD_LIBRARY_PATH