func chol
chol (X)

    Returns the cholesky decomposition, B, of X, such that B*B' = X.  B
    will be lower triangular.  X must be symmetric and positive definite.


func cols
cols (X)

    Returns the number of columns of X.  A scalar is defined to have a
    single column.


func compact
compact (X)

    Returns a compact-storage matrix whose elements are identical to X,
    which must be a symmetric matrix.  The space required by a compact
    matrix is approximately equal to the number of non-zero entries.  The
    compact representation of a scalar is itself.


func det
det (X)

    Returns the determinant of X, which must be nonsingular.  The
    determinant of a scalar is itself.


func eig
eig (X)
eig (X, V)

    Return a column vector containing the eigenvalues of X, which must be
    square.  If X is a scalar then X is returned.  If a variable V is
    specified and X is symmetric then V will contain the matrix of
    eigenvectors on output.  Otherwise, V is ignored.


func eye
eye (m)
eye (m, n)

    Returns an identity matrix of size (m x n).  If n is omitted then an
    (m x m) matrix is returned.  Both m and n must be scalars.


func inv
inv (X)

    Returns the inverse of X, or (1/X).  X must be a either a nonsingular
    matrix or a non-zero scalar.


func lu
lu (X)
lu (X, L)
lu (X, L, U)
lu (X, L, U, P)

    Computes the LU decomposition of X, which must be nonsingular.  The
    return value is row permuted combination of L and U, with the diagonal
    of L not being stored since L is unit lower triangular.  If the
    remaining parameters are variables then they will contain L, U, and/or P
    (the permutation matrix) on output, such that P*L*U=X.


func norm
norm (X)
norm (X, s)

    Returns the norm of X.  If X is a scalar then s is ignored and the
    absolute value of X is returned.  If X is a vector then s may be one of
    "1", "2", or "fro" indicating that the 1-norm, 2-norm, or frobenius-norm
    (identical to the 2-norm) is to be computed.  The default is to compute
    the 2-norm.  If X is a matrix then s may be either "1" or "fro"
    indicating that the 1-norm or frobenius-norm is to be computed.  The
    default is to compute the frobenius-norm.


func ones
ones (m)
ones (m, n)

    Returns a matrix of size (m x n) whose elements are all one.  If n is
    not specified then an (m x m) matrix is returned.  Both m and n must be
    scalars.


func qr
qr (X)
qr (X, Q)
qr (X, Q, R)

    Computes the QR decomposition of X, which must be overdetermined (tall
    and thin).  The return value is R, which is right triangular, such that
    Q'*X=R.  If a variable Q is specified then it will contain the
    orthogonal matrix of the decomposition on output.


func rand
rand ( )
rand (m)
rand (m, n)
rand (m, n, s)

    Returns a matrix of size (m x n) with randomly generated elements
    between zero and one.  If n is omitted then an (m x m) matrix is
    returned.  If both m and n are absent then a random scalar is returned.
    If s is specified and is non-zero then it used to seed then random
    number generator.  Both m and n must be scalars.


func rows
rows (X)

    Returns the number of rows of X.  A scalar is defined to have a
    single row.


func zeros func zeroes
zeros (m)
zeros (m, n)

    Returns a matrix of size (m x n) whose elements are all zero.  If n is
    not specified then an (m x m) matrix is returned.  Both m and n must be
    scalars.