Revision history [back]
 | 1 | initial version |
I'm not entirely sure what you are looking for due to the vagueness of your question, but here is a response that might help.
One way to encode a linear programming problem is:
p=MixedIntegerLinearProgram()
x=p.new_variable()
p.add_constraint(-3*x[0]+x[1]<=2)
p.add_constraint(x[1]<=11)
p.add_constraint(x[0]-x[1]<=3)
p.add_constraint(x[0]<=6)
p.set_objective(x[0]+2*x[1])
To display the objective function and constraints, use:
p.show()
To solve the LP problem, use:
p.solve()
p.get_values(x)
To display the feasible region in the 2d case, use:
p.polyhedron().show()
You can also enter the LP constraints using a matrix formulation as follows:
A=matrix([[-3,1],[0,1],[1,-1],[1,0]])
b=vector([2,11,3,6])
p=MixedIntegerLinearProgram()
x=p.new_variable()
p.set_objective(x[0]+2*x[1])
p.add_constraint(A*x<=b)
