ipo interior-point optimisation library
Variables and arrays
A convex optimisation model has variables. These are shared by the objective function and constraints. They can be bounded. For example, you may specify that some or all variables are nonnegative by setting their lower bounds to zero.
You may name a variable and generally should. The getValue() and setValue() methods get and set the value of the variable. The getValue() is important because you need it to find the values of the variables at an optimum solution.
Variables are described in ipo::Variable.
Sometimes it is inconvenient to create variables one at a time. Ipo provides an ipo::Array class to handle arrays of variables. Think of Array as like std::vector::<Variable> except that (i) an Array must be attached to a Model, (ii) an Array may have a name.
The getValue() and setValue() methods of Array let you get and set the values of the array as and from a gsl::vector.
Last modified: Tue 02 Jul 2013 04:56 pm