problem.addgencons
problem.addgencons |
Purpose
Adds one or more general constraints to the problem. Each general constraint
y = f(x1, ..., xn, c1, ..., cn) consists of one or more (input) columns xi, zero or more constant values ci and a resultant (output column) y. General constraints can be defined using operators such as
maximum and
minimum (at least one input column of any type and arbitrary number of input values),
and and
or (at least one binary input column, no constant values, binary resultant) and
absolute value (exactly one input column of arbitrary type, no constant values).
Synopsis
problem.addgencons (type, resultant, colstart, col, valstart, val)
Arguments
|
type
|
list or array containing the types of the general constraints:
|
||||||||||
|
resultant
|
Array/list containing the output variables (or indices thereof) of the general constraints.
|
||||||||||
|
colstart
|
Array/list containing the start index of each general constraint in the
col array.
|
||||||||||
|
col
|
Array/list containing the input variables in all general constraints.
|
||||||||||
|
valstart
|
Array/list containing the start index of each general constraint in the
val array (may be
None).
|
||||||||||
|
val
|
Array/list containing the constant values in all general constraints (may be
None).
|
Example
This adds two new general constraints
x2 = max(x0, x1, 5) and
x3 = |x1|:
type = [xpress.gencons_max, xpress.gencons_abs] resultant = [2, 3] colstart = [0, 2] col = [0, 1, 1] valstart = [0, 1] val = [5.0] prob.addgencons(type, resultant, colstart, col, valstart, val); prob.solve()
Further information
General constraints must be set up before solving the problem. They are converted to additional binary variables, indicator and linear constraints with the exact formulation and number of added entities depending on the performed presolving. Note that using non-binary variables in
and/
or constraints or adding constant values to them or
absolute value constraints will give an error at solve time.
Related topics