Introduction | Installation | Quick Start | Linear Regression: family = "gaussian" (default) | Commonly used function arguments | Predicting and plotting with glmnet objects | Cross-validation | Other function arguments | Linear Regression: family = "mgaussian" (multi-response) | Logistic Regression: family = "binomial" | Multinomial Regression: family = "multinomial" | Poisson Regression: family = "poisson" | Cox Regression: family = "cox" | Programmable GLM families: family = family() | Assessing models on test data | Performance measures | Prevalidation | ROC curves for binomial data | Confusion matrices for classification | Filtering variables | Other Package Features | Sparse matrix support | Fitting big and/or sparse unpenalized generalized linear models | Creating x from mixed variables and/or missing data | Progress bar | Appendix 0: Convergence Criteria | Appendix 1: Internal Parameters | Appendix 2: Comparison with Other Packages | References

1. Introduction | 2. Origins: coordinate descent and the first release (2008--2010) | 3. Filling out the matrix (2010--2018) | 4. Usability matures: v3.0 (2019) | 5. GLM extension: v4.0 (2020) | 6. Cox catches up: v4.1 (2021) | 7. Fortran → C++ port (2021--2022) | 8. Applications feeding back into the package | 9. Streamlined Cox and v5.0 (2026) | 10. Looking forward | 11. Contributors and acknowledgements | References

Abstract | The easy case: no ties | Right censored | Start / stop | Difference of cumsums | Second derivative | Probabilistic interpretation | Ties and zero weights | First derivative | A lemma or two | Derivatives revisited | Full Hessian | Non-diagonal term | Diagonal term | Appendix: Proofs of Lemmas | Proof of Lemma 1 (Forward cumsum representation) | Proof of Lemma 2 (Adjoint cumsum representation)

Introduction | Basic usage for right-censored data | Cross-validation | Handling of ties | Cox models for start-stop data | Stratified Cox models | Plotting survival curves | References

What softImpute solves | A simple example | Debiasing the fit | Using the sparse matrix version | Reduced rank SVD of large sparse matrices | Warm starts and regularization paths

Introduction to Group Lasso and Elastic Net | Single-Response Group Elastic Net | Multi-Response Group Elastic Net | Quickstart | Gaussian Group Elastic Net | Lasso | Group Lasso | GLM Group Elastic Net | Multi-Response GLM Elastic Net | Cox Models | Constraints | Non-negative Lasso | Constraints on a single group | Matrix | Some Useful Matrix Methods | Dense Matrix | Sparse Matrix | Concatenated Matrices | Mixed variables and one-hot encoding | Standardization | Extended Examples | Learning Interactions via Hierarchical Group-Lasso Regularization | Simulation Setup | Tools in adelie for fitting a glinternet model | Details: manual construction of interaction model | References

Introduction | Simple relaxed fitting | More details on relaxed fitting | Possible convergence issues for relaxed fits | Application to forward stepwise regression | References

Introduction | Using class "family" objects for the family argument | More on GLM families | Fitting Gaussian, binomial and Poisson GLMs | Timing comparisons | Fitting other GLMs | Class "glmnetfit" objects | Step size halving within iteratively reweighted least squares (IRLS)