About
What is it?
Automated Tool for Optimized Modelling (ATOM) is an open-source Python package designed to help data scientists fasten up the exploration phase of their machine learning projects. ATOM is a low-code, easy-to-use library, capable of running experiments quickly and efficiently, enabling the user to go from raw data to generating insights in just a few lines of code. Click here to get started.
What can I do with it?
ATOM is an end-to-end solution for machine learning pipelines. It supports the user from raw data ingestion to the final results' analysis. Click on the icons to read more about its main functionalities.
Who is it intended for?
- Data scientists that want to fasten up the exploration phase of their machine learning projects.
- Data scientists that want to run a simple modelling experiment without having to spend too much time on coding.
- Data scientists that are new to Python and are not (yet) familiar with all the relevant machine learning packages.
- Data analysts without extensive knowledge of machine learning that want to try out model-based solutions.
- Anyone who wants to rapidly build a Proof of Concept, for example during a hackathon.
- Anyone who is new to the field of machine learning and wants a low-code, easy to learn package, to get started building predictive pipelines.
Citing ATOM
If you use ATOM in a scientific publication, please consider citing this documentation page as the resource. ATOM’s first stable release v2.0.3 was made publicly available in November 2019. A formatted version of the citation would look like this:
ATOM v2.0.3, November 2019. URL https://tvdboom.github.io/ATOM/
BibTeX entry:
@Manual{ATOM,
title = {ATOM: A Python package for fast exploration of machine learning pipelines},
author = {Mavs},
year={2019},
mont={November},
note = {ATOM version 2.0.3},
url = {https://tvdboom.github.io/ATOM/},
}
Support
ATOM recognizes the support from JetBrains by providing core project contributors with a set of developer tools free of charge.
