problem.getmqobj
problem.getmqobj |
Purpose
Returns the nonzeros in the quadratic objective coefficients' matrix for the columns in a given range. To achieve maximum efficiency,
getmqobj returns the lower triangular part of this matrix only.
Synopsis
problem.getmqobj (mstart, mclind, dobjval, size, first, last)
Arguments
|
mstart
|
Array which will be filled with indices indicating the starting offsets in the
mclind and
dobjval arrays for each requested column. It must be length of at least
last-first+2. Column
i starts at position
mstart[i] in the
mrwind and
dmatval arrays, and has
mstart[i+1]-mstart[i] elements in it. May be
None if
size is
0.
|
|
mclind
|
Array which will be filled with at most
size column indices of the nonzero elements in the lower triangular part of
Q. May be
None if
size is
0.
|
|
dobjval
|
Array which will be filled with at most
size nonzero element values. May be
None if
size is
0.
|
|
size
|
The maximum number of elements to be returned (size of the arrays).
|
|
first
|
First column in the range.
|
|
last
|
Last column in the range.
|
Further information
The objective function is of the form
cTx+0.5xTQx where
Q is positive semi-definite for minimization problems and negative semi-definite for maximization problems. If this is not the case the optimization algorithms may converge to a local optimum or may not converge at all. Note that only the upper or lower triangular part of the
Q matrix is returned.
Related topics