The condition number of a matrix describes how sensitive it is to errors in the right-hand side vector when solving systems of equations. The factorization classes can optionally calculate an estimate of the reciprocal condition number when a factorization is constructed. When this number is near 1, the matrix is said to be well-conditioned. When the reciprocal condition number is near 0, the matrix is ill-conditioned. The solutions obtained for ill-conditioned matrices may not be very accurate. Consult a linear algebra text for more information on condition numbers.
The condition number is accessed via member function condition(). For example, if LU is an existing LU factorization, then:
double rcond = LU.condition();
returns the reciprocal condition number.
©Copyright 1999, Rogue Wave Software, Inc.
Contact Rogue Wave about documentation or support issues.