That contain all of the necessary source files, using a Makefile. TensorFlow Lite for Microcontrollers is able to generate standalone projects
You can then save a copy of theĮxample and use it as the basis of your own project. It and click hello_world to load the example. You should see anĮxample near the bottom of the list named TensorFlowLite:hello_world. Once the library has been added, go to File -> Examples.
If you are using Arduino, the Hello World example is included in theĪrduino_TensorFlowLite Arduino library, which you can download from the YouĬan obtain a version of it for your platform of choice by following the We recommend using the Hello World example as a template for new projects. Which contains build tools and their output. Which contains operation implementations and the associated code. Several other directories exist, including: These are located in a directory with the platform name, for example The build system provides for platform-specific implementations of certainįiles. See Get started with microcontrollers for a Micro_mutable_op_resolver.h to pull in only the operations your model In production applications, you should use Since all_ops_resolver.h pulls in every available operation, it Interpreter are located in the root of the project, accompanied by tests:Ĭan be used to provide the operations used by the interpreter to run the The most important files for using the TensorFlow Lite for Microcontrollers Within various embedded development environments. Pre-generated project files that provide the relevant source files in isolation Inside of the extensive TensorFlow repository, we have created scripts and Root directory has a relatively simple structure. Provides information about creating your own project. The following document outlines the basic structure of the C++ library and It is designed to be readable, easy to modify, well-tested, easy to integrate,Īnd compatible with regular TensorFlow Lite. The TensorFlow Lite for Microcontrollers C++ library is part of the