snort3/libml

LibML is a library for loading, configuring, and running machine learning models in production. It provides a simple high-level API for C++ applications. LibML uses TensorFlow with XNNPACK acceleration for low latency inference.

LibML

Description

LibML is a library for loading, configuring, and running machine learning models in production. It provides a simple high-level API for C++ applications. LibML uses TensorFlow with XNNPACK acceleration for low latency inference.

Install

./configure.sh
cd build
sudo make -j$(nproc) install

Examples

Binary Classifier

BinaryClassifier classifier;

if(!classifier.buildFromFile(model_path))
    return 1;

float output = 0.0;

if(!classifier.run(input, input_size, output))
    return 1;

std::cout << "output: " << output << "%\n";

Build Dependencies

• CMake
• C++ Compiler

libml's Issues

Hello, I'm also very interested by snortML and I'm seeking for further documentation. I looked google scholar but I couldn't find any papers nor tutorials.

          Hello, I'm also very interested by snortML and I'm seeking for further documentation. I looked google scholar but I couldn't find any papers nor tutorials.

Originally posted by @Bl4omArchie in #1 (comment)

More details about using the library

Good morning,

I am undertaking my master's degree in the area of ​​machine learning applied to intrusion detection systems such as snort. I believe it will be very useful to use this library to integrate my project with snort. Could the maintainers recommend me more sources of material so that I can carry out this development. Is there a plan to detail the documentation in the near future? Congratulations to the Talos team for the initiative in creating this library that will facilitate the addition of new features to snort.