[Previous] [Next]

XSLP_SCALE

Description
When to re-scale the SLP problem
Type
Integer
Values
0 
No re-scaling.
1 
Re-scale every SLP iteration up to XSLP_SCALECOUNT iterations after the end of barrier optimization.
2 
Re-scale every SLP iteration up to XSLP_SCALECOUNT iterations in total.
3 
Re-scale every SLP iteration until primal simplex is automatically invoked.
4 
Re-scale every SLP iteration.
5 
Re-scale every XSLP_SCALECOUNT SLP iterations.
6 
Re-scale every XSLP_SCALECOUNT SLP iterations after the end of barrier optimization.
Default value
1
Notes
During the SLP optimization, matrix entries can change considerably in magnitude, even when the formulae in the coefficients are not very nonlinear. Re-scaling of the matrix can reduce numerical errors, but may increase the time taken to achieve convergence.
Affects routines
XSLPmaxim, XSLPminim
See also
XSLP_SCALECOUNT

[Previous] [Next] [Index]
If you have any comments or suggestions about these pages, please send mail to docs@dashoptimization.com.