# Training

## Overview

This section covers steps on how to train a custom AI model using our two open-source libraries:

 * [LuxonisTrain](https://docs.luxonis.com/software-v3/ai-inference/model-source/training/luxonis-train.md) for training, and
 * [LuxonisML](https://docs.luxonis.com/software-v3/ai-inference/model-source/training/luxonis-ml.md) for support tools and helper
   functions and utilities for the Luxonis MLOps stack.

You can study each of the two individually or follow the relevant
[Tutorials](https://docs.luxonis.com/software-v3/ai-inference/model-source/training/tutorials.md) for a more hands-on guide
through the training process.

Additionally, we present here the
[DataDreamer](https://docs.luxonis.com/software-v3/ai-inference/model-source/training/datadreamer.md), a tool for synthetic
dataset generation and pre-annotation of real image datasets. This tool is especially useful for training AI models for custom
applications from scratch, not requiring any pre-existing datasets.