Titli Documentation
Titli is a comprehensive toolkit for hosting feature extraction, model training, model inference, and model evaluation of AI-based Intrusion Detection Systems (IDS).
Overview
Titli provides a modular framework for building and evaluating intrusion detection systems with a unified API. It includes:
6 IDS Models: LOF, OCSVM, VAE, Autoencoder, ICL, KitNET
Unified API: All models expose 5 consistent methods (train_model, save, load, infer, evaluate)
Efficient DataLoaders: StreamingCSVDataset for large-scale data processing
Comprehensive Evaluation: Automatic metrics computation and visualization
Easy Persistence: Simple save/load model management with default paths
Feature Extractors: Tools for extracting features from network traffic (AfterImage, NetStat)
Table of Contents
User Guide