# Model Training Methods

## What is it

## When to use it

The `AETrainer` class should be used when you have lots of images, have very
large images, or your given speed requirements are strict. Images can be
thousands of pixels tall and wide, and can have multiple channels, which makes
them very large. Encoding these images using an
{term}`autoencoder<Autoencoder>` can shrink the images significantly.

## Theory behind it

### AE Outlier detection

The encoder is trained to create dense {term}`embeddings<Embeddings>` for the
images while the decoder is trained to reconstruct the new embedding into the
original input image. The distances from the reconstructions between the test
images and original images, or the
{term}`probability distribution<Probability Distribution>` differences are used
to measure how different they are and allow for the detection of Outliers.