OmniRegress

The fast, modern Python & Rust library for all your regression adventures.

OmniRegress: A comprehensive Python & Rust library for all types of regression analysis.

🚀 Update: 4.0.0 Release!

✨ Brand New:

• 🦀 Ridge Regression (L2) — Rust implementation with L2 regularization to reduce overfitting and handle multicollinearity
• 🦀 Lasso Regression (L1) — Rust implementation with L1 regularization for automatic feature selection and sparse solutions

🔵 Basic Regression Models

• [✅] Linear Regression — Fast, pure Rust core. (Usage 🚀)
• [✅] Polynomial Regression — Nonlinear fits, Rust-powered. (Usage 🚀)
• [✅] Logistic Regression — Native Rust, robust binary classification. (Usage 🚀)
• [✅] Ridge Regression (L2) — 🛡️ Regularization to prevent overfitting.(Usage 🚀)
• [✅] Lasso Regression (L1) — ✂️ Feature selection with L1 penalty.(Usage 🚀)
• [🚧] Elastic Net — 🧬 Hybrid L1 + L2 regularization.

🟢 Specialized Regression

• Poisson Regression — 📈 For count data (e.g., website visits).
• Cox Regression — ⏳ Survival/time-to-event analysis.
• Quantile Regression — 🎯 Predicts specific percentiles (e.g., median).
• Bayesian Regression — 🎲 Incorporates prior distributions.

🟠 Nonlinear & ML-Based

• Support Vector Regression (SVR) — 🌀 Kernel magic for complex patterns.
• Decision Tree Regression — 🌳 Hierarchical, rule-based splits.
• Random Forest Regression — 🌲🌲 Ensemble of decision trees.
• Neural Network Regression — 🧠 Deep learning for high-dimensional data.

🟣 Other Advanced Types

• Gaussian Process Regression — 🔮 Probabilistic nonlinear modeling.
• Negative Binomial Regression — 🧮 Overdispersed count data.
• Multinomial Logistic Regression — 🏷️ Multi-class classification.

🧪 Testing

See the test suite: OmniRegress_test