New Features of the Beta Machine Learning Toolkit (BetaML) - Missing Value Imputation, Autoencoders, and Variable Importance Metrics

Explore the latest enhancements to BetaML.jl, a lightweight pure-Julia machine learning library, in this 11-minute conference talk from JuliaCon Local Paris 2025. Learn about BetaML's scikit-learn-inspired API design that prioritizes usability while supporting decision trees, neural networks, clustering algorithms, and essential ML workflows with simplified heuristics and one-parameter autotuning. Discover three major new features: missing value imputation techniques for handling incomplete datasets, non-linear dimensionality reduction capabilities through autoencoders, and variable importance metrics for better model interpretability. Gain insights into how this pure-Julia implementation maintains consistency across different machine learning tasks while providing practical solutions for common data science challenges.