Machine Learning Frameworks in Julia logopeterindia.net
Knowledge Hub

Machine Learning Frameworks in Julia: Flux, Lux, MLJ & More

Julia machine learning frameworks — deep learning, classical ML and wrappers, with activity notes.

15 topics•3 categories•Updated October 2026

Deep Learning in Julia

5 entries

Native Julia neural network libraries.

01Flux.jlElegant, 100% pure-Julia deep learning stack with first-class GPU support, differentiable programming, and a broad model-zoo ecosystem. 02Knet.jlKoç University's deep learning framework, supporting GPU operation and automatic differentiation using dynamic computational graphs defined in plain Julia. Last commit on the default branch was in March 2022. 03Merlin.jlA deep learning framework written in Julia, aiming to provide a fast, flexible, and compact library for building machine learning models. Last commit on the default branch was in November 2018, so it is effectively unmaintained. 04Strada.jlA deep learning library for Julia based on the popular Caffe framework, supporting convolutional and recurrent neural network training on CPU and GPU. 05Lux.jlExplicit-parameter deep learning framework for Julia, designed for composability and performance.

Classical ML & Wrappers

5 entries

Toolkits for classical machine learning and language bindings.

06ScikitLearn.jlBrings scikit-learn's uniform model interface, pipelines, evaluation, and hyperparameter tuning tools to Julia, integrating Julia- and Python-defined models. 07MLBase.jlA "Swiss knife" toolkit of functions supporting machine learning development: data preprocessing, classification scoring, performance evaluation, and cross-validation. Last commit on the default branch was in August 2023. 08MLJ.jlJulia machine learning framework with a unified interface to many models and tools for tuning and evaluation. 09TensorFlow.jlA Julia wrapper around Google's TensorFlow, bringing state-of-the-art deep learning models to Julia with an API closely mirroring the Python original. Last commit on the default branch was in August 2021. 10Apache MXNetLightweight, portable, flexible deep learning framework with dynamic dataflow scheduling, supporting Python, R, Julia, Scala, Go, and JavaScript. Apache MXNet has been retired to the Apache Attic and its GitHub repository is archived.

Related Pages

5 entries

Neighbouring pages on this site.

11Machine Learning FrameworksPython-centered ML frameworks. 12Deep Learning FrameworksDeep learning frameworks. 13Deep Learning PlatformsDeep learning platforms. 14Machine Learning LibrariesML libraries across languages. 15Machine LearningMachine learning hub.

Frequently Asked Questions

Which Julia library should I use for deep learning?

Flux.jl is the most widely used, and Lux.jl is a newer option with an explicit-parameter design; Knet.jl is an older framework.

What about classical machine learning in Julia?

MLJ.jl provides a unified interface to many models, with ScikitLearn.jl as a bridge to scikit-learn-style models.

Which projects are no longer maintained?

Merlin.jl (last commit 2018), TensorFlow.jl (2021), Knet.jl (2022) and MLBase.jl (2023) show little recent activity, and Apache MXNet has been retired.

How current is this page?

Links were checked in October 2026. Sites that block automated checks are kept and may need a manual click. Last-commit dates come from GitHub in October 2026 and show default-branch activity only; check each project's release notes before adopting it.