Neargye/hello_tf_c_api

Neural Network TensorFlow C API

What it solves

This project provides a cross-platform set of practical examples for using the TensorFlow C API. It simplifies the process of integrating TensorFlow's native libraries into C++ applications on Windows, Linux, and macOS, reducing the boilerplate required to load graphs, manage tensors, and run inference.

How it works

The project uses CMake to automate the downloading and linking of the TensorFlow 2.21.0 Python wheel's C API headers and native libraries. It includes a helper library (hello_tf_utils) that wraps raw C API calls to make them more accessible. The examples demonstrate various core operations, such as loading a GraphDef, creating tensors (including string tensors), and performing repeated inference.

Who it’s for

Developers who need to run TensorFlow models in a native C++ environment without the overhead of the Python runtime, specifically those targeting cross-platform deployment.

Highlights

  • Cross-platform support: Works on Windows, Linux, and macOS.
  • Comprehensive examples: Covers everything from basic tensor creation to image processing with OpenCV.
  • Automated dependency management: Uses CMake to fetch and configure the TensorFlow C API automatically.
  • Native performance: Enables running inference using the native C API for better efficiency in production environments.

Related

  • Project
    mattn/go-tflite

    A Go language binding for TensorFlow Lite that allows developers to run lightweight machine learning models and perform inference within Go applications.

  • Project
    tensorflow/java

    TensorFlow‑Java provides Maven‑publishable Java/JVM bindings for TensorFlow, including low‑level JNI wrappers (`tensorflow‑core`), a high‑level neural‑network API (`tensorflow‑framework`), and an independent ndarray library. It supports Linux (x86_64, arm64) and macOS (Apple Silicon) binaries, with optional GPU builds, and integrates with standard Java build tools.

  • Project
    tensorflow/tensorflow

    An end-to-end open source platform for machine learning that provides a flexible ecosystem of tools and libraries for building and deploying ML-powered applications.

  • Project
    tensorflow/cloud

    A set of APIs that allow developers to easily move TensorFlow and Keras model training from local environments to distributed training and tuning on Google Cloud Platform.