Installation¶

Here you will learn how to install the nonconform package.

Prerequisites¶

Python 3.12 or higher is recommended

Installation with pip¶

To install nonconform, run the following command in your terminal:

pip install nonconform

Optional dependencies¶

Existing optional dependencies are grouped into [data], [dev], [docs], [deep], [fdr] and [all]: - [data]: Dataset loading functionality (includes pyarrow) - [dev]: Development dependencies - [docs]: Documentation dependencies - [deep]: Deep Learning dependencies (pytorch) - [fdr]: Online False Discovery Rate control (online-fdr) - [all]: All optional dependencies

Installing with specific dependencies¶

To install with datasets support:

pip install nonconform[data]

Note: The datasets are downloaded automatically when first used and cached both in memory and on disk (in ~/.cache/nonconform/) for faster subsequent access.

To install with online FDR control for streaming scenarios:

pip install nonconform[fdr]

To install with all optional dependencies:

pip install nonconform[all]

Get started¶

You are all set to find your first anomalies!

import nonconform