Installation Guide

DataEval is a lightweight toolkit that offers powerful metric classes that can be extended through additional package installations.

This guide will show you how to install the DataEval that fits your needs!

Supported Python Versions

We currently support python versions 3.9 - 3.11

Base DataEval Packages

To keep DataEval lightweight but powerful, only the following metrics come with the base installation

Packages
Bayes Error Rate
Divergence
Upper bound Average Precision

Extras

However, DataEval also has installable extras that provide a more expansive and powerful toolkit for any user.
These extras are torch, tensorflow and all. Any extra can be installed using dataeval[extra]

Extras	Additional Packages
torch	Sufficiency
tensorflow	OOD Detection
all	Sufficiency, OOD Detection

Installation

Now that you have a chosen which DataEval to install, the following methods will show you how to install using your preferred method.
Be sure to add [extra] if you are not installing the base DataEval

pip

Installing from pip

    pip install dataeval[all]

conda-forge

Installing from conda

    conda install dataeval[all]

source

To install DataEval from source locally on Ubuntu, you will need git-lfs to download larger, binary source files and poetry for project dependency management.

    sudo apt-get install git-lfs
    pip install poetry

Pull the source down and change to the DataEval project directory.

    git clone https://github.com/aria-ml/dataeval.git
    cd dataeval

Install DataEval with optional dependencies for development.

    poetry install --all-extras --with dev

Alternatively, you can install with optional dependencies used to generate documentation as well.

    poetry install --all-extras --with dev --with docs

Now that DataEval is installed, you can run commands in the poetry virtual environment by prefixing shell commands with poetry run, or activate the virtual environment directly in the shell.

    poetry shell