@danielsimonjr/mathts-functions 0.45.0 → 0.47.0

package/dist/index.d.ts

@@ -36,10 +36,12 @@ export { newton, secant, halley } from './numeric/open-root-finders.js';
 export type { NewtonOptions, SecantOptions, HalleyOptions } from './numeric/open-root-finders.js';
 export { fsolve, root } from './numeric/fsolve.js';
 export type { FsolveOptions } from './numeric/fsolve.js';
-export { cg, minres, gmres, bicgstab } from './numeric/krylov.js';
+export { cg, minres, gmres, bicgstab, incompleteLU, incompleteCholesky } from './numeric/krylov.js';
 export type { LinearOperatorInput, Preconditioner, KrylovOptions, GmresOptions, KrylovResult, } from './numeric/krylov.js';
 export { eigsh } from './numeric/eigsh.js';
 export type { EigshOperatorInput, EigshOptions, EigshResult } from './numeric/eigsh.js';
+export { svds } from './numeric/svds.js';
+export type { SvdsOptions, SvdsResult } from './numeric/svds.js';
 export { thomasSolve, solveBanded, toeplitzSolve, ldl } from './numeric/structured-solvers.js';
 export type { LDLResult } from './numeric/structured-solvers.js';
 export { funm, cosm, sinm, complexCos, complexSin } from './numeric/matrix-functions.js';
@@ -116,4 +118,8 @@ export { mvnPdf, mvnSample } from './stats/mvn.js';
 export type { MvnVector, MvnCov, MvnSampleOptions } from './stats/mvn.js';
 export { tTestPower } from './stats/power-analysis.js';
 export type { TTestPowerAlternative, TTestPowerOptions } from './stats/power-analysis.js';
+export { gaussianProcessRegression, gpRegression } from './stats/gaussian-process.js';
+export type { GPKernel, GPOptions, GPPrediction, GPModel } from './stats/gaussian-process.js';
+export { dirichletSample, dirichletPdf, wishartSample } from './stats/multivariate-sampling.js';
+export type { SampleSeedOptions } from './stats/multivariate-sampling.js';
 //# sourceMappingURL=index.d.ts.map

package/dist/index.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"index.d.ts","sourceRoot":"","sources":["../src/index.ts"],"names":[],"mappings":"AAAA;;;;;;;;;GASG;AAGH,cAAc,kBAAkB,CAAC;AAKjC,cAAc,gBAAgB,CAAC;AAG/B,cAAc,sBAAsB,CAAC;AACrC,OAAO,EAAE,MAAM,EAAE,MAAM,iBAAiB,CAAC;AAOzC,OAAO,EAAE,EAAE,EAAE,MAAM,EAAE,MAAM,iBAAiB,CAAC;AAG7C,OAAO,EAAE,QAAQ,EAAE,WAAW,EAAE,KAAK,EAAE,MAAM,EAAE,OAAO,EAAE,QAAQ,EAAE,MAAM,yBAAyB,CAAC;AAGlG,OAAO,EAAE,IAAI,EAAE,MAAM,WAAW,CAAC;AAMjC,OAAO,EACL,iBAAiB,EACjB,YAAY,EACZ,UAAU,EACV,WAAW,EACX,aAAa,EACb,mBAAmB,EACnB,eAAe,EACf,eAAe,EACf,eAAe,GAChB,MAAM,yBAAyB,CAAC;AAGjC,OAAO,EAAE,YAAY,EAAE,oBAAoB,EAAE,UAAU,EAAE,MAAM,mBAAmB,CAAC;AACnF,YAAY,EAAE,MAAM,EAAE,MAAM,mBAAmB,CAAC;AAGhD,OAAO,EACL,KAAK,EACL,KAAK,EACL,MAAM,EACN,QAAQ,EACR,QAAQ,EACR,GAAG,EACH,GAAG,EACH,MAAM,EACN,GAAG,EACH,QAAQ,EACR,QAAQ,EACR,QAAQ,EACR,UAAU,EACV,UAAU,EACV,QAAQ,EACR,SAAS,EACT,UAAU,EACV,cAAc,EACd,GAAG,EACH,SAAS,EACT,WAAW,EACX,QAAQ,EACR,SAAS,GACV,MAAM,wBAAwB,CAAC;AAChC,YAAY,EACV,gBAAgB,EAChB,qBAAqB,EACrB,oBAAoB,EACpB,cAAc,EACd,eAAe,GAChB,MAAM,wBAAwB,CAAC;AAGhC,OAAO,EACL,KAAK,EACL,OAAO,EACP,SAAS,EACT,OAAO,EACP,OAAO,EACP,MAAM,EACN,MAAM,EACN,QAAQ,GACT,MAAM,oBAAoB,CAAC;AAG5B,OAAO,EACL,cAAc,EACd,WAAW,EACX,gBAAgB,EAChB,aAAa,EACb,kBAAkB,EAClB,IAAI,EACJ,SAAS,EACT,QAAQ,EACR,aAAa,EACb,OAAO,EACP,YAAY,EACZ,SAAS,EACT,SAAS,EACT,cAAc,EACd,UAAU,EACV,UAAU,EACV,eAAe,EACf,WAAW,EACX,WAAW,EACX,gBAAgB,GACjB,MAAM,6BAA6B,CAAC;AAGrC,OAAO,EACL,KAAK,EACL,UAAU,EACV,aAAa,EACb,QAAQ,EACR,WAAW,EACX,mBAAmB,EACnB,wBAAwB,EACxB,QAAQ,GACT,MAAM,uBAAuB,CAAC;AAC/B,YAAY,EACV,WAAW,EACX,gBAAgB,EAChB,aAAa,EACb,cAAc,EACd,iBAAiB,EACjB,eAAe,GAChB,MAAM,uBAAuB,CAAC;AAI/B,OAAO,EACL,IAAI,EACJ,IAAI,EACJ,MAAM,EACN,QAAQ,EACR,SAAS,EACT,SAAS,EACT,MAAM,EACN,eAAe,EACf,cAAc,EACd,EAAE,GACH,MAAM,mBAAmB,CAAC;AAG3B,OAAO,EAAE,OAAO,EAAE,QAAQ,EAAE,MAAM,qBAAqB,CAAC;AAKxD,OAAO,EAAE,GAAG,EAAE,MAAM,8BAA8B,CAAC;AACnD,YAAY,EAAE,SAAS,EAAE,UAAU,EAAE,MAAM,8BAA8B,CAAC;AAC1E,OAAO,EAAE,IAAI,EAAE,MAAM,uBAAuB,CAAC;AAC7C,YAAY,EAAE,WAAW,EAAE,MAAM,uBAAuB,CAAC;AAIzD,OAAO,EAAE,eAAe,EAAE,MAAM,+BAA+B,CAAC;AAChE,YAAY,EAAE,WAAW,EAAE,sBAAsB,EAAE,MAAM,+BAA+B,CAAC;AAIzF,OAAO,EAAE,MAAM,EAAE,MAAM,EAAE,MAAM,EAAE,MAAM,gCAAgC,CAAC;AACxE,YAAY,EAAE,aAAa,EAAE,aAAa,EAAE,aAAa,EAAE,MAAM,gCAAgC,CAAC;AAIlG,OAAO,EAAE,MAAM,EAAE,IAAI,EAAE,MAAM,qBAAqB,CAAC;AACnD,YAAY,EAAE,aAAa,EAAE,MAAM,qBAAqB,CAAC;AAMzD,OAAO,EAAE,EAAE,EAAE,MAAM,EAAE,KAAK,EAAE,QAAQ,EAAE,MAAM,qBAAqB,CAAC;AAClE,YAAY,EACV,mBAAmB,EACnB,cAAc,EACd,aAAa,EACb,YAAY,EACZ,YAAY,GACb,MAAM,qBAAqB,CAAC;AAM7B,OAAO,EAAE,KAAK,EAAE,MAAM,oBAAoB,CAAC;AAC3C,YAAY,EAAE,kBAAkB,EAAE,YAAY,EAAE,WAAW,EAAE,MAAM,oBAAoB,CAAC;AAOxF,OAAO,EAAE,WAAW,EAAE,WAAW,EAAE,aAAa,EAAE,GAAG,EAAE,MAAM,iCAAiC,CAAC;AAC/F,YAAY,EAAE,SAAS,EAAE,MAAM,iCAAiC,CAAC;AAQjE,OAAO,EAAE,IAAI,EAAE,IAAI,EAAE,IAAI,EAAE,UAAU,EAAE,UAAU,EAAE,MAAM,+BAA+B,CAAC;AACzF,YAAY,EACV,YAAY,EACZ,aAAa,EACb,qBAAqB,GACtB,MAAM,+BAA+B,CAAC;AAQvC,OAAO,EAAE,KAAK,EAAE,IAAI,EAAE,IAAI,EAAE,MAAM,gCAAgC,CAAC;AAKnE,OAAO,EAAE,cAAc,EAAE,MAAM,8BAA8B,CAAC;AAC9D,YAAY,EAAE,qBAAqB,EAAE,oBAAoB,EAAE,MAAM,8BAA8B,CAAC;AAKhG,OAAO,EAAE,IAAI,EAAE,MAAM,4BAA4B,CAAC;AAClD,YAAY,EAAE,WAAW,EAAE,UAAU,EAAE,MAAM,4BAA4B,CAAC;AAM1E,OAAO,EAAE,OAAO,EAAE,MAAM,sBAAsB,CAAC;AAC/C,YAAY,EAAE,YAAY,EAAE,MAAM,sBAAsB,CAAC;AAOzD,OAAO,EAAE,UAAU,EAAE,WAAW,EAAE,MAAM,sBAAsB,CAAC;AAC/D,YAAY,EAAE,iBAAiB,EAAE,YAAY,EAAE,MAAM,sBAAsB,CAAC;AAM5E,OAAO,EAAE,mBAAmB,EAAE,MAAM,0BAA0B,CAAC;AAC/D,YAAY,EAAE,KAAK,EAAE,iBAAiB,EAAE,gBAAgB,EAAE,MAAM,0BAA0B,CAAC;AAG3F,OAAO,EAAE,aAAa,EAAE,IAAI,EAAE,OAAO,EAAE,GAAG,EAAE,MAAM,uBAAuB,CAAC;AAI1E,OAAO,EAAE,IAAI,EAAE,QAAQ,EAAE,YAAY,EAAE,QAAQ,EAAE,MAAM,uBAAuB,CAAC;AAC/E,YAAY,EAAE,cAAc,EAAE,cAAc,EAAE,MAAM,uBAAuB,CAAC;AAG5E,OAAO,EAAE,gBAAgB,EAAE,MAAM,uBAAuB,CAAC;AACzD,YAAY,EAAE,gBAAgB,EAAE,MAAM,uBAAuB,CAAC;AAG9D,OAAO,EAAE,GAAG,EAAE,MAAM,aAAa,CAAC;AAClC,YAAY,EAAE,UAAU,EAAE,SAAS,EAAE,MAAM,aAAa,CAAC;AAGzD,OAAO,EAAE,KAAK,EAAE,KAAK,EAAE,UAAU,EAAE,MAAM,gCAAgC,CAAC;AAC1E,YAAY,EACV,YAAY,EACZ,wBAAwB,EACxB,2BAA2B,GAC5B,MAAM,gCAAgC,CAAC;AAGxC,OAAO,EAAE,kBAAkB,EAAE,MAAM,6BAA6B,CAAC;AACjE,YAAY,EACV,yBAAyB,EACzB,wBAAwB,GACzB,MAAM,6BAA6B,CAAC;AAGrC,OAAO,EAAE,UAAU,EAAE,eAAe,EAAE,kBAAkB,EAAE,MAAM,yBAAyB,CAAC;AAC1F,YAAY,EAAE,cAAc,EAAE,QAAQ,EAAE,MAAM,yBAAyB,CAAC;AAGxE,OAAO,EAAE,MAAM,EAAE,kBAAkB,EAAE,MAAM,uBAAuB,CAAC;AAInE,OAAO,EAAE,MAAM,EAAE,WAAW,EAAE,UAAU,EAAE,MAAM,oBAAoB,CAAC;AAIrE,OAAO,EAAE,WAAW,EAAE,MAAM,aAAa,CAAC;AAC1C,YAAY,EAAE,kBAAkB,EAAE,iBAAiB,EAAE,MAAM,aAAa,CAAC;AAIzE,OAAO,EAAE,eAAe,EAAE,YAAY,EAAE,MAAM,4BAA4B,CAAC;AAC3E,YAAY,EACV,sBAAsB,EACtB,qBAAqB,EACrB,kBAAkB,GACnB,MAAM,4BAA4B,CAAC;AAIpC,OAAO,EAAE,eAAe,EAAE,MAAM,6BAA6B,CAAC;AAC9D,YAAY,EAAE,gBAAgB,EAAE,qBAAqB,EAAE,MAAM,6BAA6B,CAAC;AAM3F,OAAO,EAAE,IAAI,EAAE,MAAM,mBAAmB,CAAC;AACzC,YAAY,EAAE,WAAW,EAAE,UAAU,EAAE,MAAM,mBAAmB,CAAC;AACjE,YAAY,EAAE,YAAY,EAAE,MAAM,uBAAuB,CAAC;AAK1D,OAAO,EAAE,IAAI,EAAE,UAAU,EAAE,MAAM,mBAAmB,CAAC;AACrD,YAAY,EACV,WAAW,EACX,UAAU,EACV,iBAAiB,EACjB,gBAAgB,GACjB,MAAM,mBAAmB,CAAC;AAG3B,OAAO,EAAE,gBAAgB,EAAE,MAAM,sBAAsB,CAAC;AAGxD,OAAO,EAAE,MAAM,EAAE,MAAM,EAAE,OAAO,EAAE,SAAS,EAAE,QAAQ,EAAE,MAAM,0BAA0B,CAAC;AAGxF,OAAO,EAAE,IAAI,EAAE,KAAK,EAAE,QAAQ,EAAE,SAAS,EAAE,OAAO,EAAE,QAAQ,EAAE,IAAI,EAAE,MAAM,yBAAyB,CAAC;AAIpG,OAAO,EAAE,MAAM,EAAE,MAAM,EAAE,KAAK,EAAE,OAAO,EAAE,OAAO,EAAE,QAAQ,EAAE,OAAO,EAAE,MAAM,wBAAwB,CAAC;AAIpG,OAAO,EACL,cAAc,EACd,KAAK,EACL,KAAK,EACL,MAAM,EACN,MAAM,EACN,UAAU,GACX,MAAM,2BAA2B,CAAC;AACnC,YAAY,EAAE,gBAAgB,EAAE,MAAM,2BAA2B,CAAC;AAClE,YAAY,EAAE,SAAS,EAAE,MAAM,4BAA4B,CAAC;AAI5D,OAAO,EAAE,IAAI,EAAE,OAAO,EAAE,OAAO,EAAE,GAAG,EAAE,MAAM,sBAAsB,CAAC;AAKnE,OAAO,EACL,SAAS,EACT,UAAU,EACV,GAAG,EACH,SAAS,EACT,IAAI,EACJ,KAAK,EACL,QAAQ,GACT,MAAM,4BAA4B,CAAC;AACpC,YAAY,EACV,gBAAgB,EAChB,UAAU,EACV,WAAW,EACX,UAAU,GACX,MAAM,4BAA4B,CAAC;AAGpC,OAAO,EACL,SAAS,EACT,eAAe,EACf,KAAK,EACL,kBAAkB,EAClB,mBAAmB,EACnB,mBAAmB,EACnB,uBAAuB,EACvB,gBAAgB,EAChB,0BAA0B,GAC3B,MAAM,qBAAqB,CAAC;AAI7B,OAAO,EACL,iBAAiB,EACjB,eAAe,EACf,iBAAiB,EACjB,aAAa,EACb,aAAa,EACb,aAAa,EACb,WAAW,EACX,UAAU,EACV,SAAS,EACT,YAAY,EACZ,aAAa,EACb,QAAQ,EACR,WAAW,GACZ,MAAM,8BAA8B,CAAC;AACtC,YAAY,EAAE,gBAAgB,EAAE,MAAM,8BAA8B,CAAC;AAGrE,OAAO,EACL,oBAAoB,EACpB,iBAAiB,EACjB,qBAAqB,GACtB,MAAM,2BAA2B,CAAC;AACnC,YAAY,EAAE,MAAM,EAAE,oBAAoB,EAAE,MAAM,2BAA2B,CAAC;AAU9E,OAAO,EACL,iBAAiB,EACjB,cAAc,EACd,cAAc,EACd,QAAQ,EACR,OAAO,EACP,OAAO,EACP,WAAW,EACX,OAAO,EACP,QAAQ,GACT,MAAM,6BAA6B,CAAC;AACrC,YAAY,EACV,eAAe,EACf,cAAc,EACd,aAAa,EACb,cAAc,GACf,MAAM,6BAA6B,CAAC;AAIrC,OAAO,EAAE,MAAM,EAAE,MAAM,EAAE,MAAM,EAAE,GAAG,EAAE,MAAM,6BAA6B,CAAC;AAK1E,OAAO,EAAE,SAAS,EAAE,QAAQ,EAAE,OAAO,EAAE,WAAW,EAAE,MAAM,kCAAkC,CAAC;AAM7F,OAAO,EAAE,QAAQ,EAAE,QAAQ,EAAE,QAAQ,EAAE,MAAM,8BAA8B,CAAC;AAC5E,OAAO,EAAE,aAAa,EAAE,MAAM,0BAA0B,CAAC;AACzD,YAAY,EAAE,mBAAmB,EAAE,MAAM,0BAA0B,CAAC;AAIpE,OAAO,EAAE,OAAO,EAAE,OAAO,EAAE,OAAO,EAAE,SAAS,EAAE,SAAS,EAAE,OAAO,EAAE,MAAM,oBAAoB,CAAC;AAM9F,OAAO,EACL,GAAG,EACH,GAAG,EACH,aAAa,EACb,WAAW,EACX,mBAAmB,EACnB,kBAAkB,GACnB,MAAM,iCAAiC,CAAC;AACzC,YAAY,EAAE,iBAAiB,EAAE,MAAM,iCAAiC,CAAC;AAKzE,OAAO,EAAE,OAAO,EAAE,MAAM,EAAE,KAAK,EAAE,SAAS,EAAE,MAAM,yBAAyB,CAAC;AAC5E,YAAY,EACV,aAAa,EACb,YAAY,EACZ,WAAW,EACX,eAAe,GAChB,MAAM,yBAAyB,CAAC;AAMjC,OAAO,EAAE,QAAQ,EAAE,QAAQ,EAAE,MAAM,uBAAuB,CAAC;AAO3D,OAAO,EACL,aAAa,EACb,SAAS,EACT,kBAAkB,EAClB,cAAc,EACd,YAAY,GACb,MAAM,+BAA+B,CAAC;AAKvC,OAAO,EAAE,GAAG,EAAE,MAAM,aAAa,CAAC;AAClC,YAAY,EAAE,SAAS,EAAE,OAAO,EAAE,UAAU,EAAE,SAAS,EAAE,MAAM,aAAa,CAAC;AAK7E,OAAO,EAAE,MAAM,EAAE,SAAS,EAAE,MAAM,gBAAgB,CAAC;AACnD,YAAY,EAAE,SAAS,EAAE,MAAM,EAAE,gBAAgB,EAAE,MAAM,gBAAgB,CAAC;AAI1E,OAAO,EAAE,UAAU,EAAE,MAAM,2BAA2B,CAAC;AACvD,YAAY,EAAE,qBAAqB,EAAE,iBAAiB,EAAE,MAAM,2BAA2B,CAAC"}
+{"version":3,"file":"index.d.ts","sourceRoot":"","sources":["../src/index.ts"],"names":[],"mappings":"AAAA;;;;;;;;;GASG;AAGH,cAAc,kBAAkB,CAAC;AAKjC,cAAc,gBAAgB,CAAC;AAG/B,cAAc,sBAAsB,CAAC;AACrC,OAAO,EAAE,MAAM,EAAE,MAAM,iBAAiB,CAAC;AAOzC,OAAO,EAAE,EAAE,EAAE,MAAM,EAAE,MAAM,iBAAiB,CAAC;AAG7C,OAAO,EAAE,QAAQ,EAAE,WAAW,EAAE,KAAK,EAAE,MAAM,EAAE,OAAO,EAAE,QAAQ,EAAE,MAAM,yBAAyB,CAAC;AAGlG,OAAO,EAAE,IAAI,EAAE,MAAM,WAAW,CAAC;AAMjC,OAAO,EACL,iBAAiB,EACjB,YAAY,EACZ,UAAU,EACV,WAAW,EACX,aAAa,EACb,mBAAmB,EACnB,eAAe,EACf,eAAe,EACf,eAAe,GAChB,MAAM,yBAAyB,CAAC;AAGjC,OAAO,EAAE,YAAY,EAAE,oBAAoB,EAAE,UAAU,EAAE,MAAM,mBAAmB,CAAC;AACnF,YAAY,EAAE,MAAM,EAAE,MAAM,mBAAmB,CAAC;AAGhD,OAAO,EACL,KAAK,EACL,KAAK,EACL,MAAM,EACN,QAAQ,EACR,QAAQ,EACR,GAAG,EACH,GAAG,EACH,MAAM,EACN,GAAG,EACH,QAAQ,EACR,QAAQ,EACR,QAAQ,EACR,UAAU,EACV,UAAU,EACV,QAAQ,EACR,SAAS,EACT,UAAU,EACV,cAAc,EACd,GAAG,EACH,SAAS,EACT,WAAW,EACX,QAAQ,EACR,SAAS,GACV,MAAM,wBAAwB,CAAC;AAChC,YAAY,EACV,gBAAgB,EAChB,qBAAqB,EACrB,oBAAoB,EACpB,cAAc,EACd,eAAe,GAChB,MAAM,wBAAwB,CAAC;AAGhC,OAAO,EACL,KAAK,EACL,OAAO,EACP,SAAS,EACT,OAAO,EACP,OAAO,EACP,MAAM,EACN,MAAM,EACN,QAAQ,GACT,MAAM,oBAAoB,CAAC;AAG5B,OAAO,EACL,cAAc,EACd,WAAW,EACX,gBAAgB,EAChB,aAAa,EACb,kBAAkB,EAClB,IAAI,EACJ,SAAS,EACT,QAAQ,EACR,aAAa,EACb,OAAO,EACP,YAAY,EACZ,SAAS,EACT,SAAS,EACT,cAAc,EACd,UAAU,EACV,UAAU,EACV,eAAe,EACf,WAAW,EACX,WAAW,EACX,gBAAgB,GACjB,MAAM,6BAA6B,CAAC;AAGrC,OAAO,EACL,KAAK,EACL,UAAU,EACV,aAAa,EACb,QAAQ,EACR,WAAW,EACX,mBAAmB,EACnB,wBAAwB,EACxB,QAAQ,GACT,MAAM,uBAAuB,CAAC;AAC/B,YAAY,EACV,WAAW,EACX,gBAAgB,EAChB,aAAa,EACb,cAAc,EACd,iBAAiB,EACjB,eAAe,GAChB,MAAM,uBAAuB,CAAC;AAI/B,OAAO,EACL,IAAI,EACJ,IAAI,EACJ,MAAM,EACN,QAAQ,EACR,SAAS,EACT,SAAS,EACT,MAAM,EACN,eAAe,EACf,cAAc,EACd,EAAE,GACH,MAAM,mBAAmB,CAAC;AAG3B,OAAO,EAAE,OAAO,EAAE,QAAQ,EAAE,MAAM,qBAAqB,CAAC;AAKxD,OAAO,EAAE,GAAG,EAAE,MAAM,8BAA8B,CAAC;AACnD,YAAY,EAAE,SAAS,EAAE,UAAU,EAAE,MAAM,8BAA8B,CAAC;AAC1E,OAAO,EAAE,IAAI,EAAE,MAAM,uBAAuB,CAAC;AAC7C,YAAY,EAAE,WAAW,EAAE,MAAM,uBAAuB,CAAC;AAIzD,OAAO,EAAE,eAAe,EAAE,MAAM,+BAA+B,CAAC;AAChE,YAAY,EAAE,WAAW,EAAE,sBAAsB,EAAE,MAAM,+BAA+B,CAAC;AAIzF,OAAO,EAAE,MAAM,EAAE,MAAM,EAAE,MAAM,EAAE,MAAM,gCAAgC,CAAC;AACxE,YAAY,EAAE,aAAa,EAAE,aAAa,EAAE,aAAa,EAAE,MAAM,gCAAgC,CAAC;AAIlG,OAAO,EAAE,MAAM,EAAE,IAAI,EAAE,MAAM,qBAAqB,CAAC;AACnD,YAAY,EAAE,aAAa,EAAE,MAAM,qBAAqB,CAAC;AAMzD,OAAO,EAAE,EAAE,EAAE,MAAM,EAAE,KAAK,EAAE,QAAQ,EAAE,YAAY,EAAE,kBAAkB,EAAE,MAAM,qBAAqB,CAAC;AACpG,YAAY,EACV,mBAAmB,EACnB,cAAc,EACd,aAAa,EACb,YAAY,EACZ,YAAY,GACb,MAAM,qBAAqB,CAAC;AAM7B,OAAO,EAAE,KAAK,EAAE,MAAM,oBAAoB,CAAC;AAC3C,YAAY,EAAE,kBAAkB,EAAE,YAAY,EAAE,WAAW,EAAE,MAAM,oBAAoB,CAAC;AAKxF,OAAO,EAAE,IAAI,EAAE,MAAM,mBAAmB,CAAC;AACzC,YAAY,EAAE,WAAW,EAAE,UAAU,EAAE,MAAM,mBAAmB,CAAC;AAOjE,OAAO,EAAE,WAAW,EAAE,WAAW,EAAE,aAAa,EAAE,GAAG,EAAE,MAAM,iCAAiC,CAAC;AAC/F,YAAY,EAAE,SAAS,EAAE,MAAM,iCAAiC,CAAC;AAQjE,OAAO,EAAE,IAAI,EAAE,IAAI,EAAE,IAAI,EAAE,UAAU,EAAE,UAAU,EAAE,MAAM,+BAA+B,CAAC;AACzF,YAAY,EACV,YAAY,EACZ,aAAa,EACb,qBAAqB,GACtB,MAAM,+BAA+B,CAAC;AAQvC,OAAO,EAAE,KAAK,EAAE,IAAI,EAAE,IAAI,EAAE,MAAM,gCAAgC,CAAC;AAKnE,OAAO,EAAE,cAAc,EAAE,MAAM,8BAA8B,CAAC;AAC9D,YAAY,EAAE,qBAAqB,EAAE,oBAAoB,EAAE,MAAM,8BAA8B,CAAC;AAKhG,OAAO,EAAE,IAAI,EAAE,MAAM,4BAA4B,CAAC;AAClD,YAAY,EAAE,WAAW,EAAE,UAAU,EAAE,MAAM,4BAA4B,CAAC;AAM1E,OAAO,EAAE,OAAO,EAAE,MAAM,sBAAsB,CAAC;AAC/C,YAAY,EAAE,YAAY,EAAE,MAAM,sBAAsB,CAAC;AAOzD,OAAO,EAAE,UAAU,EAAE,WAAW,EAAE,MAAM,sBAAsB,CAAC;AAC/D,YAAY,EAAE,iBAAiB,EAAE,YAAY,EAAE,MAAM,sBAAsB,CAAC;AAM5E,OAAO,EAAE,mBAAmB,EAAE,MAAM,0BAA0B,CAAC;AAC/D,YAAY,EAAE,KAAK,EAAE,iBAAiB,EAAE,gBAAgB,EAAE,MAAM,0BAA0B,CAAC;AAG3F,OAAO,EAAE,aAAa,EAAE,IAAI,EAAE,OAAO,EAAE,GAAG,EAAE,MAAM,uBAAuB,CAAC;AAI1E,OAAO,EAAE,IAAI,EAAE,QAAQ,EAAE,YAAY,EAAE,QAAQ,EAAE,MAAM,uBAAuB,CAAC;AAC/E,YAAY,EAAE,cAAc,EAAE,cAAc,EAAE,MAAM,uBAAuB,CAAC;AAG5E,OAAO,EAAE,gBAAgB,EAAE,MAAM,uBAAuB,CAAC;AACzD,YAAY,EAAE,gBAAgB,EAAE,MAAM,uBAAuB,CAAC;AAG9D,OAAO,EAAE,GAAG,EAAE,MAAM,aAAa,CAAC;AAClC,YAAY,EAAE,UAAU,EAAE,SAAS,EAAE,MAAM,aAAa,CAAC;AAGzD,OAAO,EAAE,KAAK,EAAE,KAAK,EAAE,UAAU,EAAE,MAAM,gCAAgC,CAAC;AAC1E,YAAY,EACV,YAAY,EACZ,wBAAwB,EACxB,2BAA2B,GAC5B,MAAM,gCAAgC,CAAC;AAGxC,OAAO,EAAE,kBAAkB,EAAE,MAAM,6BAA6B,CAAC;AACjE,YAAY,EACV,yBAAyB,EACzB,wBAAwB,GACzB,MAAM,6BAA6B,CAAC;AAGrC,OAAO,EAAE,UAAU,EAAE,eAAe,EAAE,kBAAkB,EAAE,MAAM,yBAAyB,CAAC;AAC1F,YAAY,EAAE,cAAc,EAAE,QAAQ,EAAE,MAAM,yBAAyB,CAAC;AAGxE,OAAO,EAAE,MAAM,EAAE,kBAAkB,EAAE,MAAM,uBAAuB,CAAC;AAInE,OAAO,EAAE,MAAM,EAAE,WAAW,EAAE,UAAU,EAAE,MAAM,oBAAoB,CAAC;AAIrE,OAAO,EAAE,WAAW,EAAE,MAAM,aAAa,CAAC;AAC1C,YAAY,EAAE,kBAAkB,EAAE,iBAAiB,EAAE,MAAM,aAAa,CAAC;AAIzE,OAAO,EAAE,eAAe,EAAE,YAAY,EAAE,MAAM,4BAA4B,CAAC;AAC3E,YAAY,EACV,sBAAsB,EACtB,qBAAqB,EACrB,kBAAkB,GACnB,MAAM,4BAA4B,CAAC;AAIpC,OAAO,EAAE,eAAe,EAAE,MAAM,6BAA6B,CAAC;AAC9D,YAAY,EAAE,gBAAgB,EAAE,qBAAqB,EAAE,MAAM,6BAA6B,CAAC;AAM3F,OAAO,EAAE,IAAI,EAAE,MAAM,mBAAmB,CAAC;AACzC,YAAY,EAAE,WAAW,EAAE,UAAU,EAAE,MAAM,mBAAmB,CAAC;AACjE,YAAY,EAAE,YAAY,EAAE,MAAM,uBAAuB,CAAC;AAK1D,OAAO,EAAE,IAAI,EAAE,UAAU,EAAE,MAAM,mBAAmB,CAAC;AACrD,YAAY,EACV,WAAW,EACX,UAAU,EACV,iBAAiB,EACjB,gBAAgB,GACjB,MAAM,mBAAmB,CAAC;AAG3B,OAAO,EAAE,gBAAgB,EAAE,MAAM,sBAAsB,CAAC;AAGxD,OAAO,EAAE,MAAM,EAAE,MAAM,EAAE,OAAO,EAAE,SAAS,EAAE,QAAQ,EAAE,MAAM,0BAA0B,CAAC;AAGxF,OAAO,EAAE,IAAI,EAAE,KAAK,EAAE,QAAQ,EAAE,SAAS,EAAE,OAAO,EAAE,QAAQ,EAAE,IAAI,EAAE,MAAM,yBAAyB,CAAC;AAIpG,OAAO,EAAE,MAAM,EAAE,MAAM,EAAE,KAAK,EAAE,OAAO,EAAE,OAAO,EAAE,QAAQ,EAAE,OAAO,EAAE,MAAM,wBAAwB,CAAC;AAIpG,OAAO,EACL,cAAc,EACd,KAAK,EACL,KAAK,EACL,MAAM,EACN,MAAM,EACN,UAAU,GACX,MAAM,2BAA2B,CAAC;AACnC,YAAY,EAAE,gBAAgB,EAAE,MAAM,2BAA2B,CAAC;AAClE,YAAY,EAAE,SAAS,EAAE,MAAM,4BAA4B,CAAC;AAI5D,OAAO,EAAE,IAAI,EAAE,OAAO,EAAE,OAAO,EAAE,GAAG,EAAE,MAAM,sBAAsB,CAAC;AAKnE,OAAO,EACL,SAAS,EACT,UAAU,EACV,GAAG,EACH,SAAS,EACT,IAAI,EACJ,KAAK,EACL,QAAQ,GACT,MAAM,4BAA4B,CAAC;AACpC,YAAY,EACV,gBAAgB,EAChB,UAAU,EACV,WAAW,EACX,UAAU,GACX,MAAM,4BAA4B,CAAC;AAGpC,OAAO,EACL,SAAS,EACT,eAAe,EACf,KAAK,EACL,kBAAkB,EAClB,mBAAmB,EACnB,mBAAmB,EACnB,uBAAuB,EACvB,gBAAgB,EAChB,0BAA0B,GAC3B,MAAM,qBAAqB,CAAC;AAI7B,OAAO,EACL,iBAAiB,EACjB,eAAe,EACf,iBAAiB,EACjB,aAAa,EACb,aAAa,EACb,aAAa,EACb,WAAW,EACX,UAAU,EACV,SAAS,EACT,YAAY,EACZ,aAAa,EACb,QAAQ,EACR,WAAW,GACZ,MAAM,8BAA8B,CAAC;AACtC,YAAY,EAAE,gBAAgB,EAAE,MAAM,8BAA8B,CAAC;AAGrE,OAAO,EACL,oBAAoB,EACpB,iBAAiB,EACjB,qBAAqB,GACtB,MAAM,2BAA2B,CAAC;AACnC,YAAY,EAAE,MAAM,EAAE,oBAAoB,EAAE,MAAM,2BAA2B,CAAC;AAU9E,OAAO,EACL,iBAAiB,EACjB,cAAc,EACd,cAAc,EACd,QAAQ,EACR,OAAO,EACP,OAAO,EACP,WAAW,EACX,OAAO,EACP,QAAQ,GACT,MAAM,6BAA6B,CAAC;AACrC,YAAY,EACV,eAAe,EACf,cAAc,EACd,aAAa,EACb,cAAc,GACf,MAAM,6BAA6B,CAAC;AAIrC,OAAO,EAAE,MAAM,EAAE,MAAM,EAAE,MAAM,EAAE,GAAG,EAAE,MAAM,6BAA6B,CAAC;AAK1E,OAAO,EAAE,SAAS,EAAE,QAAQ,EAAE,OAAO,EAAE,WAAW,EAAE,MAAM,kCAAkC,CAAC;AAM7F,OAAO,EAAE,QAAQ,EAAE,QAAQ,EAAE,QAAQ,EAAE,MAAM,8BAA8B,CAAC;AAC5E,OAAO,EAAE,aAAa,EAAE,MAAM,0BAA0B,CAAC;AACzD,YAAY,EAAE,mBAAmB,EAAE,MAAM,0BAA0B,CAAC;AAIpE,OAAO,EAAE,OAAO,EAAE,OAAO,EAAE,OAAO,EAAE,SAAS,EAAE,SAAS,EAAE,OAAO,EAAE,MAAM,oBAAoB,CAAC;AAM9F,OAAO,EACL,GAAG,EACH,GAAG,EACH,aAAa,EACb,WAAW,EACX,mBAAmB,EACnB,kBAAkB,GACnB,MAAM,iCAAiC,CAAC;AACzC,YAAY,EAAE,iBAAiB,EAAE,MAAM,iCAAiC,CAAC;AAKzE,OAAO,EAAE,OAAO,EAAE,MAAM,EAAE,KAAK,EAAE,SAAS,EAAE,MAAM,yBAAyB,CAAC;AAC5E,YAAY,EACV,aAAa,EACb,YAAY,EACZ,WAAW,EACX,eAAe,GAChB,MAAM,yBAAyB,CAAC;AAMjC,OAAO,EAAE,QAAQ,EAAE,QAAQ,EAAE,MAAM,uBAAuB,CAAC;AAO3D,OAAO,EACL,aAAa,EACb,SAAS,EACT,kBAAkB,EAClB,cAAc,EACd,YAAY,GACb,MAAM,+BAA+B,CAAC;AAKvC,OAAO,EAAE,GAAG,EAAE,MAAM,aAAa,CAAC;AAClC,YAAY,EAAE,SAAS,EAAE,OAAO,EAAE,UAAU,EAAE,SAAS,EAAE,MAAM,aAAa,CAAC;AAK7E,OAAO,EAAE,MAAM,EAAE,SAAS,EAAE,MAAM,gBAAgB,CAAC;AACnD,YAAY,EAAE,SAAS,EAAE,MAAM,EAAE,gBAAgB,EAAE,MAAM,gBAAgB,CAAC;AAI1E,OAAO,EAAE,UAAU,EAAE,MAAM,2BAA2B,CAAC;AACvD,YAAY,EAAE,qBAAqB,EAAE,iBAAiB,EAAE,MAAM,2BAA2B,CAAC;AAI1F,OAAO,EAAE,yBAAyB,EAAE,YAAY,EAAE,MAAM,6BAA6B,CAAC;AACtF,YAAY,EAAE,QAAQ,EAAE,SAAS,EAAE,YAAY,EAAE,OAAO,EAAE,MAAM,6BAA6B,CAAC;AAI9F,OAAO,EAAE,eAAe,EAAE,YAAY,EAAE,aAAa,EAAE,MAAM,kCAAkC,CAAC;AAChG,YAAY,EAAE,iBAAiB,EAAE,MAAM,kCAAkC,CAAC"}

package/dist/index.js

@@ -15178,6 +15178,34 @@ async function eigenvectorCentrality(adj, opts) {
 
 // src/typed/dist-objects.ts
 import { computePool as computePool9 } from "@danielsimonjr/mathts-parallel";
+
+// src/probability/util/gammaSample.ts
+function normalSampleRng(rng) {
+  const u1 = rng();
+  const u2 = rng();
+  return Math.sqrt(-2 * Math.log(u1 < 1e-300 ? 1e-300 : u1)) * Math.cos(2 * Math.PI * u2);
+}
+function gammaSampleRng(alpha, rng) {
+  if (alpha < 1) {
+    return gammaSampleRng(alpha + 1, rng) * Math.pow(rng(), 1 / alpha);
+  }
+  const d = alpha - 1 / 3;
+  const c = 1 / Math.sqrt(9 * d);
+  for (; ; ) {
+    let x;
+    let v;
+    do {
+      x = normalSampleRng(rng);
+      v = 1 + c * x;
+    } while (v <= 0);
+    v = v * v * v;
+    const u = rng();
+    if (u < 1 - 0.0331 * (x * x) * (x * x)) return d * v;
+    if (Math.log(u) < 0.5 * x * x + d * (1 - v + Math.log(v))) return d * v;
+  }
+}
+
+// src/typed/dist-objects.ts
 var DIST_WORKER_THRESHOLD = 1e5;
 var SQRT_2PI = Math.sqrt(2 * Math.PI);
 function _normalCdfStd(z) {
@@ -15425,49 +15453,13 @@ function betaDist(alpha, beta_) {
   };
 }
 function _gammaRandom(alpha) {
-  if (alpha < 1) {
-    return _gammaRandom(alpha + 1) * Math.pow(Math.random(), 1 / alpha);
-  }
-  const d = alpha - 1 / 3;
-  const c = 1 / Math.sqrt(9 * d);
-  while (true) {
-    let x;
-    let v;
-    do {
-      x = _normalRandom();
-      v = 1 + c * x;
-    } while (v <= 0);
-    v = v * v * v;
-    const u = Math.random();
-    if (u < 1 - 0.0331 * (x * x) * (x * x)) return d * v;
-    if (Math.log(u) < 0.5 * x * x + d * (1 - v + Math.log(v))) return d * v;
-  }
+  return gammaSampleRng(alpha, Math.random);
 }
 function _normalRandom() {
-  const u1 = Math.random();
-  const u2 = Math.random();
-  return Math.sqrt(-2 * Math.log(u1)) * Math.cos(2 * Math.PI * u2);
+  return normalSampleRng(Math.random);
 }
 function _gammaRandomRng(alpha, rng) {
-  if (alpha < 1) {
-    return _gammaRandomRng(alpha + 1, rng) * Math.pow(rng(), 1 / alpha);
-  }
-  const d = alpha - 1 / 3;
-  const c = 1 / Math.sqrt(9 * d);
-  while (true) {
-    let x;
-    let v;
-    do {
-      const u1 = rng();
-      const u2 = rng();
-      x = Math.sqrt(-2 * Math.log(u1 < 1e-300 ? 1e-300 : u1)) * Math.cos(2 * Math.PI * u2);
-      v = 1 + c * x;
-    } while (v <= 0);
-    v = v * v * v;
-    const u = rng();
-    if (u < 1 - 0.0331 * (x * x) * (x * x)) return d * v;
-    if (Math.log(u) < 0.5 * x * x + d * (1 - v + Math.log(v))) return d * v;
-  }
+  return gammaSampleRng(alpha, rng);
 }
 function binomialDist(n, p) {
   if (!Number.isInteger(n) || n < 0)
@@ -45090,54 +45082,15 @@ function tukeyHSD(groups, alpha = 0.05) {
 }
 
 // src/linalg-extra.ts
-import { DenseMatrix as DenseMatrix6, lu as lu3 } from "@danielsimonjr/mathts-matrix";
+import { DenseMatrix as DenseMatrix6, lu as lu3, matrixSchur as matrixSchur2 } from "@danielsimonjr/mathts-matrix";
 var _inv = inv;
 var _eigs = eigs;
 var _multiply = multiply;
 var _qr = qr;
 var transposeArr = (A) => A[0].map((_, j) => A.map((r) => r[j]));
-var identityArr = (n) => Array.from({ length: n }, (_, i) => Array.from({ length: n }, (_2, j) => i === j ? 1 : 0));
 function realSchur(M) {
-  const n = M.length;
-  let A = M.map((r) => r.slice());
-  let U = identityArr(n);
-  let m = n;
-  const normM = Math.max(...M.map((r) => r.reduce((s, v) => s + Math.abs(v), 0)), 1e-300);
-  for (let iter = 0; iter < 8e3 && m > 1; iter++) {
-    if (Math.abs(A[m - 1][m - 2]) < 1e-14 * (Math.abs(A[m - 2][m - 2]) + Math.abs(A[m - 1][m - 1]) || normM)) {
-      A[m - 1][m - 2] = 0;
-      m--;
-      continue;
-    }
-    const a = A[m - 2][m - 2];
-    const b = A[m - 2][m - 1];
-    const c = A[m - 1][m - 2];
-    const d = A[m - 1][m - 1];
-    const delta = (a - d) / 2;
-    const disc = delta * delta + b * c;
-    if (disc < 0) {
-      if (m === 2) break;
-      if (Math.abs(A[m - 2][m - 3]) < 1e-14 * (Math.abs(A[m - 3][m - 3]) + Math.abs(a) || normM)) {
-        A[m - 2][m - 3] = 0;
-        m -= 2;
-        continue;
-      }
-    }
-    const denom = Math.abs(delta) + Math.sqrt(Math.abs(disc));
-    const s = disc >= 0 && denom > 1e-300 ? d - (Math.sign(delta) || 1) * (b * c) / denom : d;
-    const As = A.map((r, i) => r.map((v, j) => v - (i === j ? s : 0)));
-    const { Q: Q2, R } = _qr(As);
-    A = _multiply(R, Q2).map((r, i) => r.map((v, j) => v + (i === j ? s : 0)));
-    U = _multiply(U, Q2);
-  }
-  for (let i = 2; i < n; i++) {
-    for (let j = 0; j < i - 1; j++) {
-      if (Math.abs(A[i][j]) > 1e-8 * (1 + Math.abs(A[i][i]) + Math.abs(A[j][j]))) {
-        throw new Error("realSchur: QR iteration failed to converge to (quasi-)triangular form");
-      }
-    }
-  }
-  return { U, S: A };
+  const { Q: Q2, T } = matrixSchur2(DenseMatrix6.fromArray(M));
+  return { U: Q2.toArray(), S: T.toArray() };
 }
 var arr3 = (x) => Array.isArray(x) ? x : Array.from(x);
 function generalizedEig(A, B) {
@@ -45476,9 +45429,14 @@ function resolvePreconditioner(a, preconditioner) {
   if (typeof preconditioner === "function") return preconditioner;
   if (!isDenseMatrix2(a)) {
     throw new Error(
-      "krylov: 'jacobi' preconditioner requires a dense matrix (need the diagonal) \u2014 pass a custom preconditioner function when using a matvec operator"
+      `krylov: '${preconditioner}' preconditioner requires a dense matrix (it reads A's entries) \u2014 pass a custom preconditioner function when using a matvec operator`
     );
   }
+  if (preconditioner === "jacobi") return makeJacobi(a);
+  if (preconditioner === "ilu") return makeILU0(a);
+  return makeIC0(a);
+}
+function makeJacobi(a) {
   const n = a.length;
   const invDiag = new Array(n);
   for (let i = 0; i < n; i++) {
@@ -45492,6 +45450,102 @@ function resolvePreconditioner(a, preconditioner) {
   }
   return (r) => r.map((v, i) => v * invDiag[i]);
 }
+function sparsityPattern(a) {
+  return a.map((row2, i) => row2.map((v, j) => v !== 0 || i === j));
+}
+function incompleteLU(a) {
+  const n = a.length;
+  const pattern = sparsityPattern(a);
+  const LU = a.map((row2) => row2.slice());
+  for (let i = 0; i < n; i++) {
+    for (let k = 0; k < i; k++) {
+      if (!pattern[i][k]) continue;
+      if (Math.abs(LU[k][k]) < 1e-300) {
+        throw new Error(`incompleteLU: ILU(0) zero pivot at U[${k}][${k}] \u2014 matrix needs pivoting`);
+      }
+      const lik = LU[i][k] / LU[k][k];
+      LU[i][k] = lik;
+      for (let j = k + 1; j < n; j++) {
+        if (!pattern[i][j]) continue;
+        LU[i][j] -= lik * LU[k][j];
+      }
+    }
+    if (Math.abs(LU[i][i]) < 1e-300) {
+      throw new Error(`incompleteLU: ILU(0) zero pivot at U[${i}][${i}] \u2014 matrix needs pivoting`);
+    }
+  }
+  const L = Array.from(
+    { length: n },
+    (_, i) => Array.from({ length: n }, (_2, j) => j < i ? LU[i][j] : j === i ? 1 : 0)
+  );
+  const U = Array.from(
+    { length: n },
+    (_, i) => Array.from({ length: n }, (_2, j) => j >= i ? LU[i][j] : 0)
+  );
+  return { L, U };
+}
+function incompleteCholesky(a) {
+  const n = a.length;
+  const src = a;
+  const pattern = sparsityPattern(src);
+  const L = Array.from({ length: n }, () => new Array(n).fill(0));
+  for (let i = 0; i < n; i++) {
+    for (let j = 0; j <= i; j++) {
+      if (!pattern[i][j]) continue;
+      let sum3 = src[i][j];
+      for (let k = 0; k < j; k++) sum3 -= L[i][k] * L[j][k];
+      if (i === j) {
+        if (sum3 <= 0) {
+          throw new Error(
+            `incompleteCholesky: IC(0) requires a symmetric positive-definite matrix (non-positive pivot ${sum3} at ${i}) \u2014 use ILU(0) for indefinite/nonsymmetric A`
+          );
+        }
+        L[i][i] = Math.sqrt(sum3);
+      } else {
+        L[i][j] = sum3 / L[j][j];
+      }
+    }
+  }
+  return { L };
+}
+function makeILU0(a) {
+  const { L, U } = incompleteLU(a);
+  const n = a.length;
+  return (r) => {
+    const y = new Array(n).fill(0);
+    for (let i = 0; i < n; i++) {
+      let s = r[i];
+      for (let j = 0; j < i; j++) s -= L[i][j] * y[j];
+      y[i] = s;
+    }
+    const x = new Array(n).fill(0);
+    for (let i = n - 1; i >= 0; i--) {
+      let s = y[i];
+      for (let j = i + 1; j < n; j++) s -= U[i][j] * x[j];
+      x[i] = s / U[i][i];
+    }
+    return x;
+  };
+}
+function makeIC0(a) {
+  const { L } = incompleteCholesky(a);
+  const n = a.length;
+  return (r) => {
+    const y = new Array(n).fill(0);
+    for (let i = 0; i < n; i++) {
+      let s = r[i];
+      for (let j = 0; j < i; j++) s -= L[i][j] * y[j];
+      y[i] = s / L[i][i];
+    }
+    const x = new Array(n).fill(0);
+    for (let i = n - 1; i >= 0; i--) {
+      let s = y[i];
+      for (let j = i + 1; j < n; j++) s -= L[j][i] * x[j];
+      x[i] = s / L[i][i];
+    }
+    return x;
+  };
+}
 function defaultMaxIter(n) {
   return Math.min(n * 10, 1e3);
 }
@@ -45545,103 +45599,55 @@ function minres(a, b, opts) {
   const tol = opts?.tol ?? 1e-10;
   const maxIter = opts?.maxIter ?? defaultMaxIter(n);
   const bNorm = norm25(b);
-  const x0 = opts?.x0 ? opts.x0.slice() : zeros2(n);
-  const r0 = subtract2(b, matvec2(x0));
-  let residual = relativeResidualNorm(norm25(r0), bNorm);
-  if (residual < tol) {
-    return { x: x0, iterations: 0, converged: true, residual };
-  }
-  const z0 = applyM(r0);
-  const beta1 = Math.sqrt(dot2(r0, z0));
-  if (beta1 < 1e-300) {
-    return { x: x0, iterations: 0, converged: residual < tol, residual };
-  }
-  const V = [zeros2(n), scale(1 / beta1, r0)];
-  const Z = [zeros2(n), scale(1 / beta1, z0)];
-  const alphas = [];
-  const betas = [beta1];
-  let bestX = x0.slice();
-  let bestResidual = residual;
+  const x = opts?.x0 ? opts.x0.slice() : zeros2(n);
+  let r1 = subtract2(b, matvec2(x));
+  let y = applyM(r1);
+  const beta1 = Math.sqrt(Math.max(0, dot2(r1, y)));
+  const residual0 = relativeResidualNorm(norm25(r1), bNorm);
+  if (residual0 < tol || beta1 < 1e-300) {
+    return { x, iterations: 0, converged: residual0 < tol, residual: residual0 };
+  }
+  let oldb = 0;
+  let beta2 = beta1;
+  let dbar = 0;
+  let epsln = 0;
+  let phibar = beta1;
+  let cs = -1;
+  let sn = 0;
+  let w = zeros2(n);
+  let w2 = zeros2(n);
+  let r2 = r1.slice();
   let iterations = 0;
-  for (let k = 1; k <= maxIter; k++) {
-    iterations = k;
-    const vK = V[k];
-    const zK = Z[k];
-    const vKm1 = V[k - 1];
-    const betaK = betas[k - 1];
-    let p = matvec2(zK);
-    p = subtract2(p, scale(betaK, vKm1));
-    const alphaK = dot2(p, zK);
-    alphas.push(alphaK);
-    p = subtract2(p, scale(alphaK, vK));
-    const zRaw = applyM(p);
-    const betaNext = Math.sqrt(Math.max(0, dot2(p, zRaw)));
-    const T = Array.from({ length: k + 1 }, () => new Array(k).fill(0));
-    for (let i = 0; i < k; i++) {
-      T[i][i] = alphas[i];
-      if (i + 1 < k) {
-        T[i + 1][i] = betas[i + 1];
-        T[i][i + 1] = betas[i + 1];
-      }
-    }
-    T[k][k - 1] = betaNext;
-    const rhs = new Array(k + 1).fill(0);
-    rhs[0] = beta1;
-    const y = solveLeastSquaresQR(T, rhs);
-    let xCandidate = x0.slice();
-    for (let i = 0; i < k; i++) {
-      xCandidate = axpy(y[i], Z[i + 1], xCandidate);
-    }
-    residual = relativeResidual(matvec2, xCandidate, b, bNorm);
-    if (residual < bestResidual) {
-      bestResidual = residual;
-      bestX = xCandidate;
-    }
-    if (residual < tol) {
-      return { x: xCandidate, iterations: k, converged: true, residual };
-    }
-    if (betaNext < 1e-300) break;
-    betas.push(betaNext);
-    V.push(scale(1 / betaNext, p));
-    Z.push(scale(1 / betaNext, zRaw));
-  }
-  return { x: bestX, iterations, converged: bestResidual < tol, residual: bestResidual };
-}
-function solveLeastSquaresQR(T, rhs) {
-  const m = T.length;
-  const n = T[0]?.length ?? 0;
-  const R = T.map((row2) => row2.slice());
-  const c = rhs.slice();
-  for (let k = 0; k < n; k++) {
-    let normX = 0;
-    for (let i = k; i < m; i++) normX += R[i][k] * R[i][k];
-    normX = Math.sqrt(normX);
-    if (normX < 1e-300) continue;
-    const alpha = R[k][k] >= 0 ? -normX : normX;
-    const v = new Array(m).fill(0);
-    v[k] = R[k][k] - alpha;
-    for (let i = k + 1; i < m; i++) v[i] = R[i][k];
-    let vNormSq = 0;
-    for (let i = k; i < m; i++) vNormSq += v[i] * v[i];
-    if (vNormSq < 1e-300) continue;
-    for (let j = k; j < n; j++) {
-      let dotVR = 0;
-      for (let i = k; i < m; i++) dotVR += v[i] * R[i][j];
-      const factor2 = 2 * dotVR / vNormSq;
-      for (let i = k; i < m; i++) R[i][j] -= factor2 * v[i];
-    }
-    let dotVC = 0;
-    for (let i = k; i < m; i++) dotVC += v[i] * c[i];
-    const factorC = 2 * dotVC / vNormSq;
-    for (let i = k; i < m; i++) c[i] -= factorC * v[i];
-  }
-  const y = new Array(n).fill(0);
-  for (let i = n - 1; i >= 0; i--) {
-    let sum3 = c[i];
-    for (let j = i + 1; j < n; j++) sum3 -= R[i][j] * y[j];
-    y[i] = Math.abs(R[i][i]) > 1e-300 ? sum3 / R[i][i] : 0;
-  }
-  return y;
+  for (let iter = 1; iter <= maxIter; iter++) {
+    iterations = iter;
+    const v = scale(1 / beta2, y);
+    y = matvec2(v);
+    if (iter >= 2) y = axpy(-beta2 / oldb, r1, y);
+    const alfa = dot2(v, y);
+    y = axpy(-alfa / beta2, r2, y);
+    r1 = r2;
+    r2 = y;
+    y = applyM(r2);
+    oldb = beta2;
+    beta2 = Math.sqrt(Math.max(0, dot2(r2, y)));
+    const oldeps = epsln;
+    const delta = cs * dbar + sn * alfa;
+    const gbar = sn * dbar - cs * alfa;
+    epsln = sn * beta2;
+    dbar = -cs * beta2;
+    const gamma2 = Math.max(Math.sqrt(gbar * gbar + beta2 * beta2), 1e-300);
+    cs = gbar / gamma2;
+    sn = beta2 / gamma2;
+    const phi = cs * phibar;
+    phibar = sn * phibar;
+    const w1 = w2;
+    w2 = w;
+    w = scale(1 / gamma2, subtract2(subtract2(v, scale(oldeps, w1)), scale(delta, w2)));
+    for (let i = 0; i < n; i++) x[i] += phi * w[i];
+    if (phibar / beta1 < tol || beta2 < 1e-300) break;
+  }
+  const residual = relativeResidual(matvec2, x, b, bNorm);
+  return { x, iterations, converged: residual < tol, residual };
 }
 function gmres(a, b, opts) {
   const n = b.length;
@@ -45939,6 +45945,67 @@ function eigsh(a, k = 1, opts) {
   return { eigenvalues, eigenvectors };
 }
 
+// src/numeric/svds.ts
+function matvecA(A, x) {
+  return A.map((row2) => {
+    let s = 0;
+    for (let j = 0; j < row2.length; j++) s += row2[j] * x[j];
+    return s;
+  });
+}
+function matvecAt(A, y, n) {
+  const out = new Array(n).fill(0);
+  for (let i = 0; i < A.length; i++) {
+    const yi = y[i];
+    const row2 = A[i];
+    for (let j = 0; j < n; j++) out[j] += row2[j] * yi;
+  }
+  return out;
+}
+function norm27(v) {
+  let s = 0;
+  for (let i = 0; i < v.length; i++) s += v[i] * v[i];
+  return Math.sqrt(s);
+}
+function svds(A, k = 1, opts) {
+  const m = A.length;
+  const n = A[0]?.length ?? 0;
+  if (m === 0 || n === 0 || A.some((row2) => row2.length !== n)) {
+    throw new Error("svds: A must be a non-empty rectangular matrix");
+  }
+  const minDim = Math.min(m, n);
+  if (!Number.isInteger(k) || k < 1 || k > minDim) {
+    throw new Error(`svds: k must be an integer between 1 and min(m,n)=${minDim}, got ${k}`);
+  }
+  const tol = opts?.tol ?? 1e-10;
+  const useAtA = m >= n;
+  const dim = useAtA ? n : m;
+  const normalOp = useAtA ? (x) => matvecAt(A, matvecA(A, x), n) : (x) => matvecA(A, matvecAt(A, x, n));
+  const { eigenvalues, eigenvectors } = eigsh(normalOp, k, {
+    which: "LM",
+    n: dim,
+    tol,
+    maxIter: opts?.maxIter
+  });
+  const s = eigenvalues.map((lambda) => Math.sqrt(Math.max(0, lambda)));
+  const primaryDim = dim;
+  const otherDim = useAtA ? m : n;
+  const primary = Array.from(
+    { length: primaryDim },
+    () => new Array(k).fill(0)
+  );
+  const other = Array.from({ length: otherDim }, () => new Array(k).fill(0));
+  for (let col = 0; col < k; col++) {
+    const pvec = eigenvectors.map((row2) => row2[col]);
+    for (let i = 0; i < primaryDim; i++) primary[i][col] = pvec[i];
+    let ovec = useAtA ? matvecA(A, pvec) : matvecAt(A, pvec, n);
+    const nrm = norm27(ovec);
+    ovec = nrm > 1e-300 ? ovec.map((v) => v / nrm) : ovec;
+    for (let i = 0; i < otherDim; i++) other[i][col] = ovec[i];
+  }
+  return useAtA ? { U: other, s, V: primary } : { U: primary, s, V: other };
+}
+
 // src/numeric/structured-solvers.ts
 function thomasSolve(sub2, diag2, sup, d) {
   const n = diag2.length;
@@ -47618,7 +47685,7 @@ function levenbergMarquardt(residual, x0, opts = {}) {
   let x = Array.from(x0);
   let lambda = 1e-3;
   const h = 1e-7;
-  const norm27 = (v) => v.reduce((s, c) => s + c * c, 0);
+  const norm28 = (v) => v.reduce((s, c) => s + c * c, 0);
   const jacobian2 = (xc, r0) => {
     const m = r0.length;
     const J = Array.from({ length: m }, () => new Array(n).fill(0));
@@ -47631,7 +47698,7 @@ function levenbergMarquardt(residual, x0, opts = {}) {
     return J;
   };
   let r = residual(x);
-  let cost = norm27(r);
+  let cost = norm28(r);
   let iter = 0;
   for (; iter < maxIter; iter++) {
     const J = jacobian2(x, r);
@@ -47662,7 +47729,7 @@ function levenbergMarquardt(residual, x0, opts = {}) {
     }
     const xNew = x.map((v, i) => v + delta[i]);
     const rNew = residual(xNew);
-    const costNew = norm27(rNew);
+    const costNew = norm28(rNew);
     if (costNew < cost) {
       x = xNew;
       r = rNew;
@@ -51959,6 +52026,261 @@ function tTestPower(effectSize, nobsOrPower, alpha, opts = {}) {
   if (!(nobs > 1)) throw new Error("tTestPower: nobs must be greater than 1");
   return powerAt(effectSize, nobs, alpha, alt);
 }
+
+// src/stats/gaussian-process.ts
+var SQRT3 = Math.sqrt(3);
+var SQRT5 = Math.sqrt(5);
+function euclidean2(a, b) {
+  let s = 0;
+  for (let i = 0; i < a.length; i++) {
+    const d = a[i] - b[i];
+    s += d * d;
+  }
+  return Math.sqrt(s);
+}
+function makeKernel(kernel, ell, sf2) {
+  switch (kernel) {
+    case "rbf":
+      return (a, b) => {
+        const r = euclidean2(a, b);
+        return sf2 * Math.exp(-(r * r) / (2 * ell * ell));
+      };
+    case "matern32":
+      return (a, b) => {
+        const u = SQRT3 * euclidean2(a, b) / ell;
+        return sf2 * (1 + u) * Math.exp(-u);
+      };
+    case "matern52":
+      return (a, b) => {
+        const r = euclidean2(a, b);
+        const u = SQRT5 * r / ell;
+        return sf2 * (1 + u + 5 * r * r / (3 * ell * ell)) * Math.exp(-u);
+      };
+    default:
+      throw new Error(`gaussianProcessRegression: unknown kernel '${String(kernel)}'`);
+  }
+}
+function forwardSolve(L, b) {
+  const n = b.length;
+  const x = new Array(n);
+  for (let i = 0; i < n; i++) {
+    let s = b[i];
+    for (let j = 0; j < i; j++) s -= L[i][j] * x[j];
+    x[i] = s / L[i][i];
+  }
+  return x;
+}
+function backSolveTranspose(L, b) {
+  const n = b.length;
+  const x = new Array(n);
+  for (let i = n - 1; i >= 0; i--) {
+    let s = b[i];
+    for (let j = i + 1; j < n; j++) s -= L[j][i] * x[j];
+    x[i] = s / L[i][i];
+  }
+  return x;
+}
+function normalizeTrainingInputs(X, y) {
+  if (!Array.isArray(X) || X.length === 0) {
+    throw new Error("gaussianProcessRegression: X must be a non-empty array of feature vectors");
+  }
+  if (y.length !== X.length) {
+    throw new Error(
+      `gaussianProcessRegression: y length ${y.length} must match number of rows ${X.length}`
+    );
+  }
+  const dim = X[0].length;
+  if (dim === 0) throw new Error("gaussianProcessRegression: feature vectors must be non-empty");
+  const Xa = X.map((row2) => {
+    if (row2.length !== dim) {
+      throw new Error("gaussianProcessRegression: all feature vectors must have the same length");
+    }
+    return [...row2];
+  });
+  return { Xa, ya: [...y], dim };
+}
+function gaussianProcessRegression(X, y, options = {}) {
+  const kernel = options.kernel ?? "rbf";
+  const ell = options.lengthScale ?? 1;
+  const sf2 = options.signalVariance ?? 1;
+  const noise = options.noise ?? 1e-10;
+  if (!(ell > 0)) throw new Error("gaussianProcessRegression: lengthScale must be positive");
+  if (!(sf2 > 0)) throw new Error("gaussianProcessRegression: signalVariance must be positive");
+  if (!(noise >= 0)) throw new Error("gaussianProcessRegression: noise must be non-negative");
+  const { Xa, ya } = normalizeTrainingInputs(X, y);
+  const n = Xa.length;
+  const k = makeKernel(kernel, ell, sf2);
+  const K = Array.from({ length: n }, () => new Array(n).fill(0));
+  for (let i = 0; i < n; i++) {
+    for (let j = i; j < n; j++) {
+      const kij = k(Xa[i], Xa[j]);
+      K[i][j] = kij;
+      K[j][i] = kij;
+    }
+    K[i][i] += noise;
+  }
+  const { L } = cholesky(K);
+  const Linvy = forwardSolve(L, ya);
+  const alphaBar = backSolveTranspose(L, Linvy);
+  let logDetHalf = 0;
+  for (let i = 0; i < n; i++) logDetHalf += Math.log(L[i][i]);
+  let yTa = 0;
+  for (let i = 0; i < n; i++) yTa += ya[i] * alphaBar[i];
+  const logMarginalLikelihood = -0.5 * yTa - logDetHalf - n / 2 * Math.log(2 * Math.PI);
+  return {
+    kernel,
+    logMarginalLikelihood,
+    predict(Xstar) {
+      const m = Xstar.length;
+      const mean7 = new Array(m);
+      const variance2 = new Array(m);
+      const std2 = new Array(m);
+      for (let t = 0; t < m; t++) {
+        const xt = Xstar[t];
+        if (xt.length !== Xa[0].length) {
+          throw new Error("gaussianProcessRegression.predict: test point dimensionality mismatch");
+        }
+        const ks = new Array(n);
+        for (let i = 0; i < n; i++) ks[i] = k(Xa[i], xt);
+        let mu = 0;
+        for (let i = 0; i < n; i++) mu += ks[i] * alphaBar[i];
+        const v = forwardSolve(L, ks);
+        let vtv = 0;
+        for (let i = 0; i < n; i++) vtv += v[i] * v[i];
+        const kss = k(xt, xt);
+        const varT = Math.max(kss - vtv, 0);
+        mean7[t] = mu;
+        variance2[t] = varT;
+        std2[t] = Math.sqrt(varT);
+      }
+      return { mean: mean7, variance: variance2, std: std2 };
+    }
+  };
+}
+var gpRegression = gaussianProcessRegression;
+
+// src/stats/multivariate-sampling.ts
+function validateAlpha(alpha, fn) {
+  if (!Array.isArray(alpha) || alpha.length < 2) {
+    throw new Error(`${fn}: alpha must be an array of at least 2 concentration parameters`);
+  }
+  for (const a of alpha) {
+    if (!(a > 0) || !Number.isFinite(a)) {
+      throw new Error(`${fn}: all concentration parameters must be positive and finite`);
+    }
+  }
+}
+function dirichletSample(alpha, n = 1, opts) {
+  validateAlpha(alpha, "dirichletSample");
+  if (!Number.isInteger(n) || n < 1) {
+    throw new Error("dirichletSample: n must be a positive integer");
+  }
+  const k = alpha.length;
+  const rng = createRng(opts?.seed ?? null);
+  const samples = new Array(n);
+  for (let s = 0; s < n; s++) {
+    const g = new Array(k);
+    let total = 0;
+    for (let i = 0; i < k; i++) {
+      const gi = gammaSampleRng(alpha[i], rng);
+      g[i] = gi;
+      total += gi;
+    }
+    const row2 = new Array(k);
+    for (let i = 0; i < k; i++) row2[i] = g[i] / total;
+    samples[s] = row2;
+  }
+  return samples;
+}
+function dirichletPdf(x, alpha) {
+  validateAlpha(alpha, "dirichletPdf");
+  if (x.length !== alpha.length) {
+    throw new Error(`dirichletPdf: x length ${x.length} must match alpha length ${alpha.length}`);
+  }
+  let sum3 = 0;
+  for (const xi of x) {
+    if (!(xi >= 0) || !(xi <= 1)) {
+      throw new Error("dirichletPdf: x must lie on the simplex (each component in [0, 1])");
+    }
+    sum3 += xi;
+  }
+  if (Math.abs(sum3 - 1) > 1e-8) {
+    throw new Error("dirichletPdf: x components must sum to 1");
+  }
+  let a0 = 0;
+  let logB = 0;
+  for (const a of alpha) {
+    logB += lgammaNumber(a);
+    a0 += a;
+  }
+  logB -= lgammaNumber(a0);
+  let logpdf = -logB;
+  for (let i = 0; i < x.length; i++) {
+    const xi = x[i];
+    const p = alpha[i] - 1;
+    if (p !== 0) {
+      logpdf += p * Math.log(xi);
+    }
+  }
+  return Math.exp(logpdf);
+}
+function validateScale(scale4, fn) {
+  if (!Array.isArray(scale4) || scale4.length === 0) {
+    throw new Error(`${fn}: scale must be a non-empty square matrix`);
+  }
+  const p = scale4.length;
+  for (const row2 of scale4) {
+    if (row2.length !== p) throw new Error(`${fn}: scale must be a ${p}x${p} square matrix`);
+  }
+  for (let i = 0; i < p; i++) {
+    for (let j = i + 1; j < p; j++) {
+      const tol = 1e-9 * (1 + Math.abs(scale4[i][j]));
+      if (Math.abs(scale4[i][j] - scale4[j][i]) > tol) {
+        throw new Error(`${fn}: scale must be symmetric`);
+      }
+    }
+  }
+  return p;
+}
+function wishartSample(df, scale4, n = 1, opts) {
+  const p = validateScale(scale4, "wishartSample");
+  if (!(df > p - 1)) {
+    throw new Error(`wishartSample: df must be greater than p\u22121 = ${p - 1}`);
+  }
+  if (!Number.isInteger(n) || n < 1) {
+    throw new Error("wishartSample: n must be a positive integer");
+  }
+  const { L } = cholesky(scale4.map((row2) => [...row2]));
+  const rng = createRng(opts?.seed ?? null);
+  const samples = new Array(n);
+  for (let s = 0; s < n; s++) {
+    const A = Array.from({ length: p }, () => new Array(p).fill(0));
+    for (let i = 0; i < p; i++) {
+      A[i][i] = Math.sqrt(2 * gammaSampleRng((df - i) / 2, rng));
+      for (let j = 0; j < i; j++) A[i][j] = normalSampleRng(rng);
+    }
+    const M = Array.from({ length: p }, () => new Array(p).fill(0));
+    for (let i = 0; i < p; i++) {
+      for (let j = 0; j <= i; j++) {
+        let sum3 = 0;
+        for (let m = j; m <= i; m++) sum3 += L[i][m] * A[m][j];
+        M[i][j] = sum3;
+      }
+    }
+    const W = Array.from({ length: p }, () => new Array(p).fill(0));
+    for (let i = 0; i < p; i++) {
+      for (let j = i; j < p; j++) {
+        let sum3 = 0;
+        const lim = Math.min(i, j);
+        for (let m = 0; m <= lim; m++) sum3 += M[i][m] * M[j][m];
+        W[i][j] = sum3;
+        W[j][i] = sum3;
+      }
+    }
+    samples[s] = W;
+  }
+  return samples;
+}
 export {
   ARRAY_WORKER_THRESHOLD,
   CAS_BATCH_THRESHOLD,
@@ -52193,6 +52515,8 @@ export {
   differences,
   digamma,
   directionalDerivative,
+  dirichletPdf,
+  dirichletSample,
   disableGpu,
   discreteLog,
   discreteUniformDist,
@@ -52382,6 +52706,7 @@ export {
   gasConstant,
   gaussQuad,
   gaussianKDE,
+  gaussianProcessRegression,
   gcd,
   gegenbauerC,
   generalizedEig,
@@ -52394,6 +52719,7 @@ export {
   gmean,
   gmres,
   goertzel,
+  gpRegression,
   gpuAdd,
   gpuMatmul,
   gpuScale,
@@ -52441,6 +52767,8 @@ export {
   ifftshift2 as ifftshift,
   im,
   implicitDiff,
+  incompleteCholesky,
+  incompleteLU,
   indexFn as index,
   indexFn,
   initializePool,
@@ -52906,6 +53234,7 @@ export {
   sum,
   summation,
   svd4 as svd,
+  svds,
   sylvester,
   symbolicEqual,
   symbolicIntegral,
@@ -52988,6 +53317,7 @@ export {
   wiener,
   wilcoxon,
   windowFunction,
+  wishartSample,
   xgcd,
   xor,
   zTransform,

package/dist/linalg-extra.d.ts

@@ -2,8 +2,12 @@ type Vec = readonly number[] | Float64Array;
 /**
  * Generalized eigenvalues of the pencil `A x = λ B x` (B nonsingular), via the
  * eigendecomposition of `B⁻¹A` — reusing `inv`, `multiply`, and the corrected
- * `eigs`. Eigenvalues match `scipy.linalg.eig(A, B)`. For ill-conditioned `B`
- * prefer a QZ solver (not yet available).
+ * `eigs` (Hessenberg + Francis double-shift). Eigenvalues match
+ * `scipy.linalg.eig(A, B)` across real, complex-pair and clustered spectra. For
+ * a *singular* or numerically near-singular `B` this squaring-free QZ formulation
+ * breaks down; extracting the pencil eigenvalues directly from the {@link qz}
+ * factors (`diag AA / diag BB`, with 2×2-block handling for complex pairs) is the
+ * future enhancement for that regime.
  */
 export declare function generalizedEig(A: readonly number[][], B: readonly number[][]): {
     values: Array<number | {
@@ -58,8 +62,10 @@ export declare function logdet(A: readonly number[][]): {
  * Generalized (QZ) Schur decomposition of the pencil `(A, B)` with `B` nonsingular:
  * returns orthogonal `Q`, `Z` and upper-(quasi-)triangular `AA`, `BB` with
  * `A = Q·AA·Zᵀ` and `B = Q·BB·Zᵀ`. Built from the real Schur of `B⁻¹A` (= Z S Zᵀ) and
- * the QR of `B·Z` (= Q·BB): then `AA = Qᵀ·A·Z`. Robust for real spectra; matches the
- * decomposition contract of `scipy.linalg.qz` (the factors are not unique).
+ * the QR of `B·Z` (= Q·BB): then `AA = Qᵀ·A·Z`. The Schur step is the hardened
+ * Hessenberg + Francis double-shift `matrixSchur` (see {@link realSchur}), so `qz`
+ * no longer stalls on non-symmetric `B⁻¹A` pencils; matches the decomposition
+ * contract of `scipy.linalg.qz` (the factors are not unique).
  */
 export declare function qz(A: readonly number[][], B: readonly number[][]): {
     AA: number[][];

package/dist/linalg-extra.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"linalg-extra.d.ts","sourceRoot":"","sources":["../src/linalg-extra.ts"],"names":[],"mappings":"AA4FA,KAAK,GAAG,GAAG,SAAS,MAAM,EAAE,GAAG,YAAY,CAAC;AAG5C;;;;;GAKG;AACH,wBAAgB,cAAc,CAC5B,CAAC,EAAE,SAAS,MAAM,EAAE,EAAE,EACtB,CAAC,EAAE,SAAS,MAAM,EAAE,EAAE,GACrB;IAAE,MAAM,EAAE,KAAK,CAAC,MAAM,GAAG;QAAE,EAAE,EAAE,MAAM,CAAC;QAAC,EAAE,EAAE,MAAM,CAAA;KAAE,CAAC,CAAC;IAAC,YAAY,CAAC,EAAE,OAAO,CAAA;CAAE,CAIhF;AAED,+EAA+E;AAC/E,wBAAgB,IAAI,CAAC,CAAC,EAAE,SAAS,MAAM,EAAE,EAAE,EAAE,CAAC,SAAI,GAAG,MAAM,EAAE,EAAE,CAE9D;AAED,+EAA+E;AAC/E,wBAAgB,IAAI,CAAC,CAAC,EAAE,SAAS,MAAM,EAAE,EAAE,EAAE,CAAC,SAAI,GAAG,MAAM,EAAE,EAAE,CAE9D;AAED;;;;GAIG;AACH,wBAAgB,MAAM,CAAC,CAAC,EAAE,GAAG,EAAE,CAAC,CAAC,EAAE,MAAM,EAAE,UAAU,UAAQ,GAAG,MAAM,EAAE,EAAE,CAMzE;AAED;;;GAGG;AACH,wBAAgB,QAAQ,CAAC,CAAC,EAAE,GAAG,EAAE,CAAC,CAAC,EAAE,GAAG,GAAG,MAAM,EAAE,EAAE,CAQpD;AAED,2EAA2E;AAC3E,wBAAgB,SAAS,CAAC,CAAC,EAAE,GAAG,GAAG,MAAM,EAAE,EAAE,CAM5C;AAED;;;;GAIG;AACH,wBAAgB,SAAS,CAAC,MAAM,EAAE,GAAG,GAAG,MAAM,EAAE,EAAE,CASjD;AAED;;;;;;GAMG;AACH,wBAAgB,eAAe,CAC7B,SAAS,EAAE,SAAS,MAAM,EAAE,EAAE,EAC9B,IAAI,GAAE;IAAE,UAAU,CAAC,EAAE,OAAO,CAAA;CAAO,GAClC,MAAM,EAAE,EAAE,CAYZ;AAED;;;;GAIG;AACH,wBAAgB,MAAM,CAAC,CAAC,EAAE,SAAS,MAAM,EAAE,EAAE,GAAG;IAAE,IAAI,EAAE,MAAM,CAAC;IAAC,KAAK,EAAE,MAAM,CAAA;CAAE,CA2B9E;AAED;;;;;;GAMG;AACH,wBAAgB,EAAE,CAChB,CAAC,EAAE,SAAS,MAAM,EAAE,EAAE,EACtB,CAAC,EAAE,SAAS,MAAM,EAAE,EAAE,GACrB;IAAE,EAAE,EAAE,MAAM,EAAE,EAAE,CAAC;IAAC,EAAE,EAAE,MAAM,EAAE,EAAE,CAAC;IAAC,CAAC,EAAE,MAAM,EAAE,EAAE,CAAC;IAAC,CAAC,EAAE,MAAM,EAAE,EAAE,CAAA;CAAE,CASlE"}
+{"version":3,"file":"linalg-extra.d.ts","sourceRoot":"","sources":["../src/linalg-extra.ts"],"names":[],"mappings":"AAuCA,KAAK,GAAG,GAAG,SAAS,MAAM,EAAE,GAAG,YAAY,CAAC;AAG5C;;;;;;;;;GASG;AACH,wBAAgB,cAAc,CAC5B,CAAC,EAAE,SAAS,MAAM,EAAE,EAAE,EACtB,CAAC,EAAE,SAAS,MAAM,EAAE,EAAE,GACrB;IAAE,MAAM,EAAE,KAAK,CAAC,MAAM,GAAG;QAAE,EAAE,EAAE,MAAM,CAAC;QAAC,EAAE,EAAE,MAAM,CAAA;KAAE,CAAC,CAAC;IAAC,YAAY,CAAC,EAAE,OAAO,CAAA;CAAE,CAIhF;AAED,+EAA+E;AAC/E,wBAAgB,IAAI,CAAC,CAAC,EAAE,SAAS,MAAM,EAAE,EAAE,EAAE,CAAC,SAAI,GAAG,MAAM,EAAE,EAAE,CAE9D;AAED,+EAA+E;AAC/E,wBAAgB,IAAI,CAAC,CAAC,EAAE,SAAS,MAAM,EAAE,EAAE,EAAE,CAAC,SAAI,GAAG,MAAM,EAAE,EAAE,CAE9D;AAED;;;;GAIG;AACH,wBAAgB,MAAM,CAAC,CAAC,EAAE,GAAG,EAAE,CAAC,CAAC,EAAE,MAAM,EAAE,UAAU,UAAQ,GAAG,MAAM,EAAE,EAAE,CAMzE;AAED;;;GAGG;AACH,wBAAgB,QAAQ,CAAC,CAAC,EAAE,GAAG,EAAE,CAAC,CAAC,EAAE,GAAG,GAAG,MAAM,EAAE,EAAE,CAQpD;AAED,2EAA2E;AAC3E,wBAAgB,SAAS,CAAC,CAAC,EAAE,GAAG,GAAG,MAAM,EAAE,EAAE,CAM5C;AAED;;;;GAIG;AACH,wBAAgB,SAAS,CAAC,MAAM,EAAE,GAAG,GAAG,MAAM,EAAE,EAAE,CASjD;AAED;;;;;;GAMG;AACH,wBAAgB,eAAe,CAC7B,SAAS,EAAE,SAAS,MAAM,EAAE,EAAE,EAC9B,IAAI,GAAE;IAAE,UAAU,CAAC,EAAE,OAAO,CAAA;CAAO,GAClC,MAAM,EAAE,EAAE,CAYZ;AAED;;;;GAIG;AACH,wBAAgB,MAAM,CAAC,CAAC,EAAE,SAAS,MAAM,EAAE,EAAE,GAAG;IAAE,IAAI,EAAE,MAAM,CAAC;IAAC,KAAK,EAAE,MAAM,CAAA;CAAE,CA2B9E;AAED;;;;;;;;GAQG;AACH,wBAAgB,EAAE,CAChB,CAAC,EAAE,SAAS,MAAM,EAAE,EAAE,EACtB,CAAC,EAAE,SAAS,MAAM,EAAE,EAAE,GACrB;IAAE,EAAE,EAAE,MAAM,EAAE,EAAE,CAAC;IAAC,EAAE,EAAE,MAAM,EAAE,EAAE,CAAC;IAAC,CAAC,EAAE,MAAM,EAAE,EAAE,CAAC;IAAC,CAAC,EAAE,MAAM,EAAE,EAAE,CAAA;CAAE,CASlE"}

package/dist/numeric/krylov.d.ts

@@ -15,8 +15,15 @@
  */
 /** A linear operator: either a dense matrix or a matvec callback `x -> A x`. */
 export type LinearOperatorInput = number[][] | ((x: number[]) => number[]);
-/** Preconditioner: `'jacobi'` (diagonal, dense-matrix only) or a custom `M⁻¹` callback. */
-export type Preconditioner = 'jacobi' | ((r: number[]) => number[]);
+/**
+ * Preconditioner. The three built-ins require a dense matrix (they read `A`'s
+ * entries): `'jacobi'` (diagonal `M⁻¹ = diag(1/Aᵢᵢ)`), `'ilu'` (ILU(0) —
+ * incomplete LU with zero fill on `A`'s sparsity pattern, for general `A`),
+ * `'ic'` (IC(0) — incomplete Cholesky, for symmetric positive-definite `A`).
+ * Or pass a custom `M⁻¹` callback (the only option that works with a
+ * matvec-only operator).
+ */
+export type Preconditioner = 'jacobi' | 'ilu' | 'ic' | ((r: number[]) => number[]);
 /** Common options accepted by every solver in this module. */
 export interface KrylovOptions {
     /** Initial guess (default: the zero vector). */
@@ -44,6 +51,31 @@ export interface KrylovResult {
     /** Final relative residual `‖b − A x‖₂ / ‖b‖₂`. */
     residual: number;
 }
+/**
+ * ILU(0) — incomplete LU factorization with zero fill. Returns unit-lower `L`
+ * and upper `U` (dense, with zeros outside `A`'s sparsity pattern) such that
+ * `(L·U)ᵢⱼ = Aᵢⱼ` on that pattern, dropping any fill that would arise outside
+ * it. The classic preconditioner for iterative solvers on general sparse `A`.
+ *
+ * @example
+ * incompleteLU([[4,1,0],[1,4,1],[0,1,4]]) // tridiagonal → exact LU (no fill)
+ */
+export declare function incompleteLU(a: readonly number[][]): {
+    L: number[][];
+    U: number[][];
+};
+/**
+ * IC(0) — incomplete Cholesky factorization with zero fill, for symmetric
+ * positive-definite `A`. Returns lower-triangular `L` on `A`'s sparsity pattern
+ * with `(L·Lᵀ)ᵢⱼ = Aᵢⱼ` there. Throws on a non-positive pivot (the `A`-is-SPD
+ * precondition failed).
+ *
+ * @example
+ * incompleteCholesky([[4,1,0],[1,4,1],[0,1,4]]) // tridiagonal → exact Cholesky
+ */
+export declare function incompleteCholesky(a: readonly number[][]): {
+    L: number[][];
+};
 /**
  * Conjugate Gradient (CG) — for symmetric positive-definite `A`.
  *
@@ -54,13 +86,19 @@ export declare function cg(a: LinearOperatorInput, b: number[], opts?: KrylovOpt
 /**
  * MINRES — for symmetric (possibly indefinite) `A`.
  *
- * Builds the preconditioned Lanczos tridiagonalization of `A` and, at each
- * step, solves the growing `(k+1) x k` tridiagonal least-squares problem
- * `min ||beta1 e1 - T_k y||` via a small dense Householder QR. This is the
- * textbook MINRES minimization restated directly (rather than the
- * incrementally-updated Givens-rotation form), which keeps it simple to
- * verify against a closed-form oracle at the cost of some redundant work
- * per iteration — acceptable since `maxIter` is bounded by `min(10n, 1000)`.
+ * The Paige–Saunders short-recurrence MINRES (Paige & Saunders 1975): a
+ * preconditioned Lanczos tridiagonalization coupled with an incrementally
+ * updated Givens-rotation QR of the tridiagonal. Each iteration does a fixed
+ * number of length-`n` vector operations (one matvec + a handful of `axpy`s)
+ * and **O(1)** scalar work — no growing least-squares is ever formed or
+ * solved. The solution is advanced through a running 3-term `w`-recurrence, so
+ * the whole solve is **O(k·n)** for `k` iterations rather than the O(k³) of the
+ * former "re-solve the growing `(k+1)×k` tridiagonal each step" formulation.
+ *
+ * With a preconditioner `M`, the loop converges the `M⁻¹`-norm relative
+ * residual `‖r_k‖_{M⁻¹} / ‖r_0‖_{M⁻¹}` (the estimate MINRES minimizes); for
+ * `M = I` this equals `‖b − A x‖₂ / ‖b‖₂`. The reported `residual`/`converged`
+ * are always computed from the true Euclidean residual (one final matvec).
  *
  * @example
  * minres([[0, 1], [1, 0]], [1, 2]) // => { x: [2, 1], converged: true, ... }

package/dist/numeric/krylov.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"krylov.d.ts","sourceRoot":"","sources":["../../src/numeric/krylov.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;GAcG;AAEH,gFAAgF;AAChF,MAAM,MAAM,mBAAmB,GAAG,MAAM,EAAE,EAAE,GAAG,CAAC,CAAC,CAAC,EAAE,MAAM,EAAE,KAAK,MAAM,EAAE,CAAC,CAAC;AAE3E,2FAA2F;AAC3F,MAAM,MAAM,cAAc,GAAG,QAAQ,GAAG,CAAC,CAAC,CAAC,EAAE,MAAM,EAAE,KAAK,MAAM,EAAE,CAAC,CAAC;AAEpE,8DAA8D;AAC9D,MAAM,WAAW,aAAa;IAC5B,gDAAgD;IAChD,EAAE,CAAC,EAAE,MAAM,EAAE,CAAC;IACd,+DAA+D;IAC/D,GAAG,CAAC,EAAE,MAAM,CAAC;IACb,wDAAwD;IACxD,OAAO,CAAC,EAAE,MAAM,CAAC;IACjB,qEAAqE;IACrE,cAAc,CAAC,EAAE,cAAc,CAAC;CACjC;AAED,4DAA4D;AAC5D,MAAM,WAAW,YAAa,SAAQ,aAAa;IACjD,mCAAmC;IACnC,OAAO,CAAC,EAAE,MAAM,CAAC;CAClB;AAED,sDAAsD;AACtD,MAAM,WAAW,YAAY;IAC3B,4BAA4B;IAC5B,CAAC,EAAE,MAAM,EAAE,CAAC;IACZ,sCAAsC;IACtC,UAAU,EAAE,MAAM,CAAC;IACnB,yDAAyD;IACzD,SAAS,EAAE,OAAO,CAAC;IACnB,mDAAmD;IACnD,QAAQ,EAAE,MAAM,CAAC;CAClB;AAmGD;;;;;GAKG;AACH,wBAAgB,EAAE,CAAC,CAAC,EAAE,mBAAmB,EAAE,CAAC,EAAE,MAAM,EAAE,EAAE,IAAI,CAAC,EAAE,aAAa,GAAG,YAAY,CAwC1F;AAED;;;;;;;;;;;;;GAaG;AACH,wBAAgB,MAAM,CAAC,CAAC,EAAE,mBAAmB,EAAE,CAAC,EAAE,MAAM,EAAE,EAAE,IAAI,CAAC,EAAE,aAAa,GAAG,YAAY,CAyF9F;AA8CD;;;;;GAKG;AACH,wBAAgB,KAAK,CAAC,CAAC,EAAE,mBAAmB,EAAE,CAAC,EAAE,MAAM,EAAE,EAAE,IAAI,CAAC,EAAE,YAAY,GAAG,YAAY,CA2F5F;AAED;;;;;GAKG;AACH,wBAAgB,QAAQ,CAAC,CAAC,EAAE,mBAAmB,EAAE,CAAC,EAAE,MAAM,EAAE,EAAE,IAAI,CAAC,EAAE,aAAa,GAAG,YAAY,CAqEhG"}
+{"version":3,"file":"krylov.d.ts","sourceRoot":"","sources":["../../src/numeric/krylov.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;GAcG;AAEH,gFAAgF;AAChF,MAAM,MAAM,mBAAmB,GAAG,MAAM,EAAE,EAAE,GAAG,CAAC,CAAC,CAAC,EAAE,MAAM,EAAE,KAAK,MAAM,EAAE,CAAC,CAAC;AAE3E;;;;;;;GAOG;AACH,MAAM,MAAM,cAAc,GAAG,QAAQ,GAAG,KAAK,GAAG,IAAI,GAAG,CAAC,CAAC,CAAC,EAAE,MAAM,EAAE,KAAK,MAAM,EAAE,CAAC,CAAC;AAEnF,8DAA8D;AAC9D,MAAM,WAAW,aAAa;IAC5B,gDAAgD;IAChD,EAAE,CAAC,EAAE,MAAM,EAAE,CAAC;IACd,+DAA+D;IAC/D,GAAG,CAAC,EAAE,MAAM,CAAC;IACb,wDAAwD;IACxD,OAAO,CAAC,EAAE,MAAM,CAAC;IACjB,qEAAqE;IACrE,cAAc,CAAC,EAAE,cAAc,CAAC;CACjC;AAED,4DAA4D;AAC5D,MAAM,WAAW,YAAa,SAAQ,aAAa;IACjD,mCAAmC;IACnC,OAAO,CAAC,EAAE,MAAM,CAAC;CAClB;AAED,sDAAsD;AACtD,MAAM,WAAW,YAAY;IAC3B,4BAA4B;IAC5B,CAAC,EAAE,MAAM,EAAE,CAAC;IACZ,sCAAsC;IACtC,UAAU,EAAE,MAAM,CAAC;IACnB,yDAAyD;IACzD,SAAS,EAAE,OAAO,CAAC;IACnB,mDAAmD;IACnD,QAAQ,EAAE,MAAM,CAAC;CAClB;AA6FD;;;;;;;;GAQG;AACH,wBAAgB,YAAY,CAAC,CAAC,EAAE,SAAS,MAAM,EAAE,EAAE,GAAG;IAAE,CAAC,EAAE,MAAM,EAAE,EAAE,CAAC;IAAC,CAAC,EAAE,MAAM,EAAE,EAAE,CAAA;CAAE,CA4BrF;AAED;;;;;;;;GAQG;AACH,wBAAgB,kBAAkB,CAAC,CAAC,EAAE,SAAS,MAAM,EAAE,EAAE,GAAG;IAAE,CAAC,EAAE,MAAM,EAAE,EAAE,CAAA;CAAE,CAuB5E;AA6DD;;;;;GAKG;AACH,wBAAgB,EAAE,CAAC,CAAC,EAAE,mBAAmB,EAAE,CAAC,EAAE,MAAM,EAAE,EAAE,IAAI,CAAC,EAAE,aAAa,GAAG,YAAY,CAwC1F;AAED;;;;;;;;;;;;;;;;;;;GAmBG;AACH,wBAAgB,MAAM,CAAC,CAAC,EAAE,mBAAmB,EAAE,CAAC,EAAE,MAAM,EAAE,EAAE,IAAI,CAAC,EAAE,aAAa,GAAG,YAAY,CA0E9F;AAED;;;;;GAKG;AACH,wBAAgB,KAAK,CAAC,CAAC,EAAE,mBAAmB,EAAE,CAAC,EAAE,MAAM,EAAE,EAAE,IAAI,CAAC,EAAE,YAAY,GAAG,YAAY,CA2F5F;AAED;;;;;GAKG;AACH,wBAAgB,QAAQ,CAAC,CAAC,EAAE,mBAAmB,EAAE,CAAC,EAAE,MAAM,EAAE,EAAE,IAAI,CAAC,EAAE,aAAa,GAAG,YAAY,CAqEhG"}

package/dist/numeric/svds.d.ts

@@ -0,0 +1,32 @@
+/** Options accepted by {@link svds}. */
+export interface SvdsOptions {
+    /** Convergence tolerance forwarded to the Lanczos eigensolver (default 1e-10). */
+    tol?: number;
+    /** Maximum Lanczos steps (default `min(max(2k + 20, 20), dim)`). */
+    maxIter?: number;
+}
+/**
+ * Result of {@link svds}. Singular values are returned **descending**
+ * (`s[0]` largest — matching this library's full {@link svd}); note this is the
+ * opposite of `scipy.sparse.linalg.svds`, which returns them ascending.
+ * Singular vectors are stored as **columns**: `U[i][j]` is the `i`-th component
+ * of the `j`-th left singular vector (for `s[j]`), and likewise `V` for the
+ * right singular vectors.
+ */
+export interface SvdsResult {
+    /** Left singular vectors, `m × k`, as columns. */
+    U: number[][];
+    /** The `k` largest singular values, descending. */
+    s: number[];
+    /** Right singular vectors, `n × k`, as columns. */
+    V: number[][];
+}
+/**
+ * The `k` largest singular triplets of `A` (dense `m × n`) via Lanczos on the
+ * smaller normal operator.
+ *
+ * @example
+ * svds([[1,2,0],[0,3,1],[4,0,2]], 2) // => { U, s: [σ₁, σ₂] (descending), V }
+ */
+export declare function svds(A: number[][], k?: number, opts?: SvdsOptions): SvdsResult;
+//# sourceMappingURL=svds.d.ts.map

package/dist/numeric/svds.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"svds.d.ts","sourceRoot":"","sources":["../../src/numeric/svds.ts"],"names":[],"mappings":"AAqBA,wCAAwC;AACxC,MAAM,WAAW,WAAW;IAC1B,kFAAkF;IAClF,GAAG,CAAC,EAAE,MAAM,CAAC;IACb,oEAAoE;IACpE,OAAO,CAAC,EAAE,MAAM,CAAC;CAClB;AAED;;;;;;;GAOG;AACH,MAAM,WAAW,UAAU;IACzB,kDAAkD;IAClD,CAAC,EAAE,MAAM,EAAE,EAAE,CAAC;IACd,mDAAmD;IACnD,CAAC,EAAE,MAAM,EAAE,CAAC;IACZ,mDAAmD;IACnD,CAAC,EAAE,MAAM,EAAE,EAAE,CAAC;CACf;AA0BD;;;;;;GAMG;AACH,wBAAgB,IAAI,CAAC,CAAC,EAAE,MAAM,EAAE,EAAE,EAAE,CAAC,SAAI,EAAE,IAAI,CAAC,EAAE,WAAW,GAAG,UAAU,CAkDzE"}

package/dist/probability/util/gammaSample.d.ts

@@ -0,0 +1,19 @@
+/**
+ * Shared RNG-driven Gamma / standard-normal variate generators.
+ *
+ * Extracted so the distribution objects (`typed/dist-objects.ts`) and the
+ * multivariate samplers (`stats/multivariate-sampling.ts`) share ONE
+ * Marsaglia & Tsang gamma implementation rather than each carrying a copy.
+ */
+/**
+ * Standard normal N(0,1) variate via Box-Muller, driven by a caller-supplied
+ * uniform PRNG (so seeded draws are reproducible). Guards `log(0)`.
+ */
+export declare function normalSampleRng(rng: () => number): number;
+/**
+ * Gamma(alpha, 1) variate (shape `alpha`, unit scale) via Marsaglia & Tsang's
+ * method, driven by a caller-supplied uniform PRNG. For `alpha < 1` it uses the
+ * boosting identity `Gamma(a) = Gamma(a+1)·U^(1/a)`.
+ */
+export declare function gammaSampleRng(alpha: number, rng: () => number): number;
+//# sourceMappingURL=gammaSample.d.ts.map

package/dist/probability/util/gammaSample.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"gammaSample.d.ts","sourceRoot":"","sources":["../../../src/probability/util/gammaSample.ts"],"names":[],"mappings":"AAAA;;;;;;GAMG;AAEH;;;GAGG;AACH,wBAAgB,eAAe,CAAC,GAAG,EAAE,MAAM,MAAM,GAAG,MAAM,CAIzD;AAED;;;;GAIG;AACH,wBAAgB,cAAc,CAAC,KAAK,EAAE,MAAM,EAAE,GAAG,EAAE,MAAM,MAAM,GAAG,MAAM,CAkBvE"}

package/dist/stats/gaussian-process.d.ts

@@ -0,0 +1,49 @@
+/** Supported covariance kernels. */
+export type GPKernel = 'rbf' | 'matern32' | 'matern52';
+/** Options for {@link gaussianProcessRegression}. */
+export interface GPOptions {
+    /** Covariance kernel. Default `'rbf'` (squared-exponential). */
+    kernel?: GPKernel;
+    /** Kernel length-scale ℓ (> 0). Default `1`. */
+    lengthScale?: number;
+    /** Signal variance σ_f² (kernel amplitude, > 0). Default `1`. */
+    signalVariance?: number;
+    /** I.i.d. Gaussian noise variance α added to the diagonal (≥ 0). Default `1e-10`. */
+    noise?: number;
+}
+/** Posterior prediction at a set of test points. */
+export interface GPPrediction {
+    /** Posterior mean at each test point. */
+    mean: number[];
+    /** Posterior variance at each test point (≥ 0). */
+    variance: number[];
+    /** Posterior standard deviation at each test point (= √variance). */
+    std: number[];
+}
+/** A fitted Gaussian-process regressor. */
+export interface GPModel {
+    /** Predict the posterior mean/variance/std at the given test points. */
+    predict(Xstar: readonly (readonly number[])[]): GPPrediction;
+    /** Log marginal likelihood log p(y | X) of the training data under the prior. */
+    logMarginalLikelihood: number;
+    /** The resolved kernel name. */
+    kernel: GPKernel;
+}
+/**
+ * Fit a Gaussian-process regressor to training points `X` (array of
+ * length-d feature vectors) and targets `y`, then return a model exposing
+ * `.predict(Xstar)` (posterior mean/variance/std) and the log marginal
+ * likelihood.
+ *
+ * @example
+ * const gp = gaussianProcessRegression(
+ *   [[-4], [-3], [-1], [0], [2]],
+ *   [-2, 0, 1, 2, -1],
+ *   { kernel: 'rbf', lengthScale: 1.2, signalVariance: 1.5, noise: 1e-2 }
+ * );
+ * gp.predict([[-0.5], [10]]); // mean ≈ [1.5872, ~0], std ≈ [0.1453, 1.2247]
+ */
+export declare function gaussianProcessRegression(X: readonly (readonly number[])[], y: readonly number[], options?: GPOptions): GPModel;
+/** Alias of {@link gaussianProcessRegression}. */
+export declare const gpRegression: typeof gaussianProcessRegression;
+//# sourceMappingURL=gaussian-process.d.ts.map

package/dist/stats/gaussian-process.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"gaussian-process.d.ts","sourceRoot":"","sources":["../../src/stats/gaussian-process.ts"],"names":[],"mappings":"AAuBA,oCAAoC;AACpC,MAAM,MAAM,QAAQ,GAAG,KAAK,GAAG,UAAU,GAAG,UAAU,CAAC;AAEvD,qDAAqD;AACrD,MAAM,WAAW,SAAS;IACxB,gEAAgE;IAChE,MAAM,CAAC,EAAE,QAAQ,CAAC;IAClB,gDAAgD;IAChD,WAAW,CAAC,EAAE,MAAM,CAAC;IACrB,iEAAiE;IACjE,cAAc,CAAC,EAAE,MAAM,CAAC;IACxB,qFAAqF;IACrF,KAAK,CAAC,EAAE,MAAM,CAAC;CAChB;AAED,oDAAoD;AACpD,MAAM,WAAW,YAAY;IAC3B,yCAAyC;IACzC,IAAI,EAAE,MAAM,EAAE,CAAC;IACf,mDAAmD;IACnD,QAAQ,EAAE,MAAM,EAAE,CAAC;IACnB,qEAAqE;IACrE,GAAG,EAAE,MAAM,EAAE,CAAC;CACf;AAED,2CAA2C;AAC3C,MAAM,WAAW,OAAO;IACtB,wEAAwE;IACxE,OAAO,CAAC,KAAK,EAAE,SAAS,CAAC,SAAS,MAAM,EAAE,CAAC,EAAE,GAAG,YAAY,CAAC;IAC7D,iFAAiF;IACjF,qBAAqB,EAAE,MAAM,CAAC;IAC9B,gCAAgC;IAChC,MAAM,EAAE,QAAQ,CAAC;CAClB;AA0FD;;;;;;;;;;;;;GAaG;AACH,wBAAgB,yBAAyB,CACvC,CAAC,EAAE,SAAS,CAAC,SAAS,MAAM,EAAE,CAAC,EAAE,EACjC,CAAC,EAAE,SAAS,MAAM,EAAE,EACpB,OAAO,GAAE,SAAc,GACtB,OAAO,CAoET;AAED,kDAAkD;AAClD,eAAO,MAAM,YAAY,kCAA4B,CAAC"}

package/dist/stats/multivariate-sampling.d.ts

@@ -0,0 +1,41 @@
+/** Options for the multivariate samplers (shared shape with mvnSample). */
+export interface SampleSeedOptions {
+    /** Seed for the deterministic RNG (reproducible draws). Omit for a
+     * time-seeded, non-reproducible generator. */
+    seed?: string | number;
+}
+/**
+ * Draw `n` samples from a Dirichlet(`alpha`) distribution via the
+ * Gamma-normalization method. Each sample is a length-k vector on the simplex
+ * (non-negative, summing to 1). Returns an `n`-length array of length-k vectors.
+ *
+ * @example
+ * dirichletSample([2, 3, 5], 10000, { seed: 42 });
+ * // empirical mean ≈ [0.2, 0.3, 0.5] = αᵢ / Σα
+ */
+export declare function dirichletSample(alpha: readonly number[], n?: number, opts?: SampleSeedOptions): number[][];
+/**
+ * Dirichlet(`alpha`) probability density at a point `x` on the simplex.
+ * `pdf(x) = (1 / B(α)) · Πᵢ xᵢ^(αᵢ−1)` with
+ * `B(α) = Πᵢ Γ(αᵢ) / Γ(Σαᵢ)`. Matches `scipy.stats.dirichlet.pdf`.
+ *
+ * @example
+ * dirichletPdf([0.2, 0.3, 0.5], [2, 3, 4]); // 7.56
+ */
+export declare function dirichletPdf(x: readonly number[], alpha: readonly number[]): number;
+/**
+ * Draw `n` samples from a Wishart(`df`, `scale`) distribution via the Bartlett
+ * decomposition: with `L = chol(scale)` and a lower-triangular `A` whose
+ * diagonal entries are `√χ²(df−i)` (0-indexed) and whose strictly-lower entries
+ * are N(0,1), each sample is `W = (L·A)(L·A)ᵀ` — a `p×p` symmetric
+ * positive-definite matrix. Returns an `n`-length array of `p×p` matrices.
+ *
+ * `df` must exceed `p − 1` (so every χ² degree-of-freedom is positive). The
+ * mean of the distribution is `df · scale`.
+ *
+ * @example
+ * wishartSample(6, [[2, 0.5], [0.5, 1]], 10000, { seed: 7 });
+ * // empirical mean ≈ [[12, 3], [3, 6]] = df · scale
+ */
+export declare function wishartSample(df: number, scale: readonly (readonly number[])[], n?: number, opts?: SampleSeedOptions): number[][][];
+//# sourceMappingURL=multivariate-sampling.d.ts.map

package/dist/stats/multivariate-sampling.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"multivariate-sampling.d.ts","sourceRoot":"","sources":["../../src/stats/multivariate-sampling.ts"],"names":[],"mappings":"AAuBA,2EAA2E;AAC3E,MAAM,WAAW,iBAAiB;IAChC;kDAC8C;IAC9C,IAAI,CAAC,EAAE,MAAM,GAAG,MAAM,CAAC;CACxB;AAaD;;;;;;;;GAQG;AACH,wBAAgB,eAAe,CAC7B,KAAK,EAAE,SAAS,MAAM,EAAE,EACxB,CAAC,SAAI,EACL,IAAI,CAAC,EAAE,iBAAiB,GACvB,MAAM,EAAE,EAAE,CAqBZ;AAED;;;;;;;GAOG;AACH,wBAAgB,YAAY,CAAC,CAAC,EAAE,SAAS,MAAM,EAAE,EAAE,KAAK,EAAE,SAAS,MAAM,EAAE,GAAG,MAAM,CAkCnF;AAqBD;;;;;;;;;;;;;GAaG;AACH,wBAAgB,aAAa,CAC3B,EAAE,EAAE,MAAM,EACV,KAAK,EAAE,SAAS,CAAC,SAAS,MAAM,EAAE,CAAC,EAAE,EACrC,CAAC,SAAI,EACL,IAAI,CAAC,EAAE,iBAAiB,GACvB,MAAM,EAAE,EAAE,EAAE,CA0Cd"}

package/dist/typed/dist-objects.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"dist-objects.d.ts","sourceRoot":"","sources":["../../src/typed/dist-objects.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;;;;GAuBG;AASH,0CAA0C;AAC1C,KAAK,GAAG,GAAG,MAAM,CAAC;AAElB;;;GAGG;AACH,eAAO,MAAM,qBAAqB,SAAU,CAAC;AAE7C;;GAEG;AACH,MAAM,WAAW,cAAc;IAC7B;;;OAGG;IACH,IAAI,CAAC,EAAE,MAAM,CAAC;IACd;;;;OAIG;IACH,WAAW,CAAC,EAAE,MAAM,CAAC;CACtB;AAED;;GAEG;AACH,MAAM,WAAW,YAAY;IAC3B,6CAA6C;IAC7C,GAAG,EAAE,CAAC,CAAC,EAAE,GAAG,KAAK,GAAG,CAAC;IACrB,uCAAuC;IACvC,GAAG,EAAE,CAAC,CAAC,EAAE,GAAG,KAAK,GAAG,CAAC;IACrB,sCAAsC;IACtC,QAAQ,EAAE,CAAC,CAAC,EAAE,GAAG,KAAK,GAAG,CAAC;IAC1B,wBAAwB;IACxB,IAAI,EAAE,GAAG,CAAC;IACV,4BAA4B;IAC5B,QAAQ,EAAE,GAAG,CAAC;IACd,oDAAoD;IACpD,MAAM,EAAE,MAAM,GAAG,CAAC;IAClB;;;;;;;;;OASG;IACH,OAAO,EAAE,CAAC,CAAC,EAAE,MAAM,EAAE,IAAI,CAAC,EAAE,cAAc,KAAK,OAAO,CAAC,YAAY,CAAC,CAAC;CACtE;AAuPD;;;;;;;;;;;;GAYG;AACH,wBAAgB,UAAU,CAAC,EAAE,GAAE,GAAO,EAAE,KAAK,GAAE,GAAO,GAAG,YAAY,CA6BpE;AAMD;;;;;;;;;;GAUG;AACH,wBAAgB,QAAQ,CAAC,KAAK,EAAE,GAAG,EAAE,KAAK,EAAE,GAAG,GAAG,YAAY,CAiD7D;AA6DD;;;;;;;;;;GAUG;AACH,wBAAgB,YAAY,CAAC,CAAC,EAAE,MAAM,EAAE,CAAC,EAAE,GAAG,GAAG,YAAY,CAsD5D;AAMD;;;;;;;;;GASG;AACH,wBAAgB,cAAc,CAAC,CAAC,EAAE,MAAM,GAAG,YAAY,CAsDtD;AAMD;;;;;;;;;GASG;AACH,wBAAgB,eAAe,CAAC,MAAM,GAAE,GAAO,GAAG,YAAY,CAuB7D;AAMD;;;;;;;;;;GAUG;AACH,wBAAgB,KAAK,CAAC,EAAE,EAAE,GAAG,EAAE,EAAE,EAAE,GAAG,GAAG,YAAY,CAwDpD;AAMD;;;;;;;;;;GAUG;AACH,wBAAgB,SAAS,CAAC,KAAK,EAAE,GAAG,EAAE,IAAI,GAAE,GAAO,GAAG,YAAY,CA8CjE;AAMD;;;;;;;;;;GAUG;AACH,wBAAgB,aAAa,CAAC,EAAE,GAAE,GAAO,EAAE,KAAK,GAAE,GAAO,GAAG,YAAY,CAiCvE;AAMD;;;;;;;;;GASG;AACH,wBAAgB,WAAW,CAAC,MAAM,EAAE,GAAG,GAAG,YAAY,CAoDrD;AAMD;;;;;;;;;GASG;AACH,wBAAgB,KAAK,CAAC,EAAE,EAAE,GAAG,GAAG,YAAY,CAmD3C;AAMD;;;;;;;;;;GAUG;AACH,wBAAgB,WAAW,CAAC,CAAC,GAAE,GAAO,EAAE,CAAC,GAAE,GAAO,GAAG,YAAY,CA2BhE;AAMD;;;;;;;;;;GAUG;AACH,wBAAgB,WAAW,CAAC,CAAC,EAAE,GAAG,EAAE,MAAM,GAAE,GAAO,GAAG,YAAY,CA8BjE;AAmCD;;;;;;;GAOG;AACH,wBAAgB,kBAAkB,CAChC,UAAU,EAAE,MAAM,EAClB,SAAS,EAAE,MAAM,EACjB,KAAK,EAAE,MAAM,GACZ,YAAY,CAmDd;AAED;;;;;;GAMG;AACH,wBAAgB,oBAAoB,CAAC,CAAC,EAAE,MAAM,EAAE,CAAC,EAAE,GAAG,GAAG,YAAY,CAwCpE;AAcD;;;GAGG;AACH,wBAAgB,UAAU,CAAC,CAAC,EAAE,MAAM,EAAE,EAAE,EAAE,MAAM,GAAG,YAAY,CAiB9D;AAED;;;GAGG;AACH,wBAAgB,YAAY,CAAC,KAAK,EAAE,MAAM,GAAG,YAAY,CAmBxD;AAED;;;;GAIG;AACH,wBAAgB,cAAc,CAAC,CAAC,EAAE,MAAM,EAAE,CAAC,EAAE,MAAM,EAAE,CAAC,EAAE,MAAM,GAAG,YAAY,CAoC5E;AAED;;;;GAIG;AACH,wBAAgB,mBAAmB,CAAC,EAAE,EAAE,MAAM,EAAE,EAAE,EAAE,MAAM,GAAG,YAAY,CA4BxE;AAED;;;;GAIG;AACH,wBAAgB,UAAU,CAAC,EAAE,EAAE,MAAM,EAAE,IAAI,EAAE,MAAM,GAAG,YAAY,CAqBjE;AAED;;;;;GAKG;AACH,wBAAgB,YAAY,CAAC,EAAE,EAAE,MAAM,EAAE,MAAM,EAAE,MAAM,GAAG,YAAY,CAyCrE;AA0BD,+DAA+D;AAC/D,MAAM,WAAW,wBAAwB;IACvC,GAAG,EAAE,CAAC,CAAC,EAAE,MAAM,EAAE,KAAK,GAAG,CAAC;IAC1B,IAAI,EAAE,MAAM,EAAE,CAAC;IACf,GAAG,EAAE,MAAM,EAAE,EAAE,CAAC;CACjB;AAED;;;;;;GAMG;AACH,wBAAgB,kBAAkB,CAAC,IAAI,EAAE,MAAM,EAAE,EAAE,GAAG,EAAE,MAAM,EAAE,EAAE,GAAG,wBAAwB,CAyB5F"}
+{"version":3,"file":"dist-objects.d.ts","sourceRoot":"","sources":["../../src/typed/dist-objects.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;;;;GAuBG;AAUH,0CAA0C;AAC1C,KAAK,GAAG,GAAG,MAAM,CAAC;AAElB;;;GAGG;AACH,eAAO,MAAM,qBAAqB,SAAU,CAAC;AAE7C;;GAEG;AACH,MAAM,WAAW,cAAc;IAC7B;;;OAGG;IACH,IAAI,CAAC,EAAE,MAAM,CAAC;IACd;;;;OAIG;IACH,WAAW,CAAC,EAAE,MAAM,CAAC;CACtB;AAED;;GAEG;AACH,MAAM,WAAW,YAAY;IAC3B,6CAA6C;IAC7C,GAAG,EAAE,CAAC,CAAC,EAAE,GAAG,KAAK,GAAG,CAAC;IACrB,uCAAuC;IACvC,GAAG,EAAE,CAAC,CAAC,EAAE,GAAG,KAAK,GAAG,CAAC;IACrB,sCAAsC;IACtC,QAAQ,EAAE,CAAC,CAAC,EAAE,GAAG,KAAK,GAAG,CAAC;IAC1B,wBAAwB;IACxB,IAAI,EAAE,GAAG,CAAC;IACV,4BAA4B;IAC5B,QAAQ,EAAE,GAAG,CAAC;IACd,oDAAoD;IACpD,MAAM,EAAE,MAAM,GAAG,CAAC;IAClB;;;;;;;;;OASG;IACH,OAAO,EAAE,CAAC,CAAC,EAAE,MAAM,EAAE,IAAI,CAAC,EAAE,cAAc,KAAK,OAAO,CAAC,YAAY,CAAC,CAAC;CACtE;AAuPD;;;;;;;;;;;;GAYG;AACH,wBAAgB,UAAU,CAAC,EAAE,GAAE,GAAO,EAAE,KAAK,GAAE,GAAO,GAAG,YAAY,CA6BpE;AAMD;;;;;;;;;;GAUG;AACH,wBAAgB,QAAQ,CAAC,KAAK,EAAE,GAAG,EAAE,KAAK,EAAE,GAAG,GAAG,YAAY,CAiD7D;AA0BD;;;;;;;;;;GAUG;AACH,wBAAgB,YAAY,CAAC,CAAC,EAAE,MAAM,EAAE,CAAC,EAAE,GAAG,GAAG,YAAY,CAsD5D;AAMD;;;;;;;;;GASG;AACH,wBAAgB,cAAc,CAAC,CAAC,EAAE,MAAM,GAAG,YAAY,CAsDtD;AAMD;;;;;;;;;GASG;AACH,wBAAgB,eAAe,CAAC,MAAM,GAAE,GAAO,GAAG,YAAY,CAuB7D;AAMD;;;;;;;;;;GAUG;AACH,wBAAgB,KAAK,CAAC,EAAE,EAAE,GAAG,EAAE,EAAE,EAAE,GAAG,GAAG,YAAY,CAwDpD;AAMD;;;;;;;;;;GAUG;AACH,wBAAgB,SAAS,CAAC,KAAK,EAAE,GAAG,EAAE,IAAI,GAAE,GAAO,GAAG,YAAY,CA8CjE;AAMD;;;;;;;;;;GAUG;AACH,wBAAgB,aAAa,CAAC,EAAE,GAAE,GAAO,EAAE,KAAK,GAAE,GAAO,GAAG,YAAY,CAiCvE;AAMD;;;;;;;;;GASG;AACH,wBAAgB,WAAW,CAAC,MAAM,EAAE,GAAG,GAAG,YAAY,CAoDrD;AAMD;;;;;;;;;GASG;AACH,wBAAgB,KAAK,CAAC,EAAE,EAAE,GAAG,GAAG,YAAY,CAmD3C;AAMD;;;;;;;;;;GAUG;AACH,wBAAgB,WAAW,CAAC,CAAC,GAAE,GAAO,EAAE,CAAC,GAAE,GAAO,GAAG,YAAY,CA2BhE;AAMD;;;;;;;;;;GAUG;AACH,wBAAgB,WAAW,CAAC,CAAC,EAAE,GAAG,EAAE,MAAM,GAAE,GAAO,GAAG,YAAY,CA8BjE;AAmCD;;;;;;;GAOG;AACH,wBAAgB,kBAAkB,CAChC,UAAU,EAAE,MAAM,EAClB,SAAS,EAAE,MAAM,EACjB,KAAK,EAAE,MAAM,GACZ,YAAY,CAmDd;AAED;;;;;;GAMG;AACH,wBAAgB,oBAAoB,CAAC,CAAC,EAAE,MAAM,EAAE,CAAC,EAAE,GAAG,GAAG,YAAY,CAwCpE;AAcD;;;GAGG;AACH,wBAAgB,UAAU,CAAC,CAAC,EAAE,MAAM,EAAE,EAAE,EAAE,MAAM,GAAG,YAAY,CAiB9D;AAED;;;GAGG;AACH,wBAAgB,YAAY,CAAC,KAAK,EAAE,MAAM,GAAG,YAAY,CAmBxD;AAED;;;;GAIG;AACH,wBAAgB,cAAc,CAAC,CAAC,EAAE,MAAM,EAAE,CAAC,EAAE,MAAM,EAAE,CAAC,EAAE,MAAM,GAAG,YAAY,CAoC5E;AAED;;;;GAIG;AACH,wBAAgB,mBAAmB,CAAC,EAAE,EAAE,MAAM,EAAE,EAAE,EAAE,MAAM,GAAG,YAAY,CA4BxE;AAED;;;;GAIG;AACH,wBAAgB,UAAU,CAAC,EAAE,EAAE,MAAM,EAAE,IAAI,EAAE,MAAM,GAAG,YAAY,CAqBjE;AAED;;;;;GAKG;AACH,wBAAgB,YAAY,CAAC,EAAE,EAAE,MAAM,EAAE,MAAM,EAAE,MAAM,GAAG,YAAY,CAyCrE;AA0BD,+DAA+D;AAC/D,MAAM,WAAW,wBAAwB;IACvC,GAAG,EAAE,CAAC,CAAC,EAAE,MAAM,EAAE,KAAK,GAAG,CAAC;IAC1B,IAAI,EAAE,MAAM,EAAE,CAAC;IACf,GAAG,EAAE,MAAM,EAAE,EAAE,CAAC;CACjB;AAED;;;;;;GAMG;AACH,wBAAgB,kBAAkB,CAAC,IAAI,EAAE,MAAM,EAAE,EAAE,GAAG,EAAE,MAAM,EAAE,EAAE,GAAG,wBAAwB,CAyB5F"}

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@danielsimonjr/mathts-functions",
-  "version": "0.45.0",
+  "version": "0.47.0",
   "description": "Mathematical functions for MathTS - arithmetic, algebra, trigonometry, statistics, and more",
   "author": "Daniel Simon Jr.",
   "license": "MIT",