[P] eqx-learn: Classical machine learning using JAX and Equinox

2026-02-16 · Source: Machine Learning · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Software Development & Engineering · Depth: Intermediate, quick

Summary

eqx-learn is a new Python library designed for classical machine learning models, built on JAX and Equinox, aiming to provide a scikit-learn-like API within the JAX ecosystem. Developed to address the lack of Equinox-native classical ML tools, it currently supports models such as Principal Component Analysis (PCA) and Gaussian Process Regression (GPR). The library's design emphasizes a decoupled approach where models advertise capabilities via specific methods (e.g., `solve(X, y)`, `condition(X, y)`, `loss()`), and a generic `fit` function handles training. This structure is intended to align with the JAX functional programming style and offers potential for extensibility.

Key takeaway

For research scientists requiring classical machine learning models within the JAX/Equinox ecosystem, eqx-learn offers a direct, scikit-learn-inspired solution. You should explore this library if your work involves PCA or Gaussian Process Regression and you need an Equinox-native implementation, potentially contributing additional models or fitting algorithms to expand its utility.

Key insights

eqx-learn offers a JAX-based, Equinox-native library for classical ML models with a scikit-learn-like API.

Principles

• Decouple model capabilities from fitting algorithms.
• Embrace mixed OO/functional style for ML models.

Method

Models implement specific methods (e.g., `solve`, `condition`, `loss`) to advertise capabilities, then a free `fit` function trains them accordingly.

In practice

• Use eqx-learn for PCA and Gaussian Process Regression.
• Integrate Equinox Modules into JAX-based ML workflows.

Topics

• JAX
• Equinox
• Classical Machine Learning
• scikit-learn API
• Gaussian Process Regression

Code references

• eqx-learn/eqx-learn
• paramrf/paramrf

Best for: AI Scientist, Research Scientist, AI Engineer, Machine Learning Engineer, AI Researcher

Related on AIssential

• Parax v0.5: Parametric Modeling in JAX [P] — Parax v0.5 offers flexible, opt-in parametric modeling tools for JAX with an extensible API.
• The Future of AI on the Web: How TensorFlow Powers Browser-Based Machine Learning — TensorFlow.js enables client-side machine learning in browsers, enhancing privacy, speed, and scalability for web applications.
• STELLAR: Spatio-Temporal Environmental Learning with Latent Alignment and Refinement for Long-Tailed Species Distribution Modeling — STELLAR jointly optimizes dynamic habitat context and community structure in a shared latent space for improved species distribution modeling.
• Olmix: A Framework for Data Mixing Throughout LM Development — Olmix efficiently manages evolving data mixtures in LM development through empirical study and mixture reuse.
• Dynestyx: A Probabilistic Programming Library for Dynamical Systems — Dynestyx unifies state-space model specification and Bayesian inference, making advanced methods accessible in probabilistic programming.
• XMSE-Aware Adaptive Empirical Bayes Estimation — An XMSE-aware mixed estimator adaptively combines ML and Empirical Bayes to mitigate kernel misspecification risks.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Machine Learning.