You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Tracking issue: #391 — Improve test coverage significantly (staged plan).
Scope
Fill the untested substantive files and revive disabled tests. Highest-ROI stage.
Tasks
src/linalg/traits/high_order.rs: implement the empty /* TODO: Add tests */ mod tests. Cover all 4 transpose-flag branches of ab ((true,true), (false,true), (true,false), (false,false)) and non-square inputs.
src/linear/lasso_optimizer.rs: add direct tests for InteriorPointOptimizer (currently only exercised transitively via lasso.rs::lasso_fit_predict). Cover new, optimize on a small l1-regularized least-squares problem, convergence on a known-answer input.
src/error/mod.rs: tests for all Failed constructors (fit, predict, transform, because), every FailedError variant, Display formatting, PartialEq.
src/api.rs: trait-coverage tests for UnsupervisedEstimator / SupervisedEstimator / Predictor / Transformer / NoParameters wiring where behavior is testable.
src/rand_custom.rs: RNG determinism tests across std_rand (feature) vs SmallRng paths, and seed reproducibility.
Revive the 13 commented-out #[test]s across 12 files (mostly disabled serde round-trip tests in decomposition/pca.rs, decomposition/svd.rs, svm/svr.rs, linear/*, dataset/*, svm/search/svc_params.rs). Gate them with #[cfg(feature = "serde")].
Tracking issue: #391 — Improve test coverage significantly (staged plan).
Scope
Fill the untested substantive files and revive disabled tests. Highest-ROI stage.
Tasks
src/linalg/traits/high_order.rs: implement the empty/* TODO: Add tests */ mod tests. Cover all 4 transpose-flag branches ofab((true,true),(false,true),(true,false),(false,false)) and non-square inputs.src/linear/lasso_optimizer.rs: add direct tests forInteriorPointOptimizer(currently only exercised transitively vialasso.rs::lasso_fit_predict). Covernew,optimizeon a small l1-regularized least-squares problem, convergence on a known-answer input.src/error/mod.rs: tests for allFailedconstructors (fit,predict,transform,because), everyFailedErrorvariant,Displayformatting,PartialEq.src/api.rs: trait-coverage tests forUnsupervisedEstimator/SupervisedEstimator/Predictor/Transformer/NoParameterswiring where behavior is testable.src/rand_custom.rs: RNG determinism tests acrossstd_rand(feature) vsSmallRngpaths, and seed reproducibility.#[test]s across 12 files (mostly disabledserderound-trip tests indecomposition/pca.rs,decomposition/svd.rs,svm/svr.rs,linear/*,dataset/*,svm/search/svc_params.rs). Gate them with#[cfg(feature = "serde")].src/dataset/diabetes.rs::boston_dataset→diabetes_datasetsrc/algorithm/sort/quick_sort.rs::with_capacity→quick_argsort(copy-paste fromheap_select.rs)Verification
cargo test --all-featurescargo fmt --all -- --checkcargo clippy --all-features -- -Drust-2018-idioms -DwarningsDo you want to work on this issue?
yes — PR will target
development.