SeqMetrics: a unified library for performance metrics calculation in Python

Published in Journal of Open Source Software, 2024

Recommended citation: Rubab, F., Iftikhar, S., & Abbas, A. (2024). SeqMetrics: a unified library for performance metrics calculation in Python. Journal of Open Source Software, 9(99), 6450. https://doi.org/10.21105/joss.06450

The SeqMetrics library is a new Python tool designed to help people evaluate how well their machine learning models are performing. Unlike existing tools that offer only a few options or require complex setups, SeqMetrics includes over 130 easy-to-use metrics for both regression and classification tasks, making it suitable for a wide range of fields like hydrology, finance, and engineering. It works with both tabular and time series data, and it’s simple to use whether you prefer writing code or using a graphical user interface. The library is efficient, easy to install, and doesn’t rely on many other packages, which makes it beginner-friendly and ideal for users who want a fast, reliable, and privacy-conscious way to check their models’ accuracy.

Download paper here