Non-negative least squares¶
A non-negative least squares problem seeks a solution to the problem
\[\min_x \| A z - y \|_2 \text{ such that } z \ge 0.\]
The following routines attempt to solve such problems by recognizing the equivalence to the quadratic program
\[\min_z \frac{1}{2} z^T P z + q^T z \text{ such that } z \ge 0\]
with \(P = A^T A\) and \(q = -A^T y\).
C++ API¶
-
Int
NonNegativeLeastSquares(const Matrix<Real> &A, const Matrix<Real> &Y, Matrix<Real> &Z, Real rho = 1, Real alpha = 1.2, Int maxIter = 500, Real absTol = 1e-6, Real relTol = 1e-4, bool inv = true, bool progress = true)¶
-
Int
NonNegativeLeastSquares(const AbstractDistMatrix<Real> &A, const AbstractDistMatrix<Real> &Y, AbstractDistMatrix<Real> &Z, Real rho = 1, Real alpha = 1.2, Int maxIter = 500, Real absTol = 1e-6, Real relTol = 1e-4, bool inv = true, bool progress = true)¶
C API¶
-
ElError
ElNonNegativeLeastSquares_s(ElConstMatrix_s A, ElConstMatrix_s Y, ElMatrix_s Z, ElInt* numIts)¶
-
ElError
ElNonNegativeLeastSquares_d(ElConstMatrix_d A, ElConstMatrix_d Y, ElMatrix_d Z, ElInt* numIts)¶