Branch and Bound algorithm

Hello :)

I'm currently working on implementing the Branch and Bound algorithm with Python.
My values are binary integers.

For implementing I'm using a CSV. In this CSV my program is able to get the values of the Objective function and also my constraints.

After this I implement the variables depending on the number of rows in my CSV.
This leads to my problem. I want to create an algorithm. This means that my program needs to be adjusted depending on the number of variables and accordingly I can not use the expression "x1_value = milp_model.getVarByName("x1").x" and so on. Because in some kind of cases I have two variables and therefore I only need x1_value and x2_value and in other cases i have four variables and the programm needs to adapt to this.
Is there any other option to get the value of my variable? I thought about something like this:

def update_values(n, milp_model):
    values = []
    for i in range(1, n+1):
        variable_name = f"x{i}"
        value = milp_model.getVarByName(variable_name).x
        values.append(value)
    return values

but in this case I have the problem that variable_name must be enclosed in quotation marks, so this doesn't work.

I would appreciate any help!

Kind regards
Chiara

David Torres Sanchez

Gurobi Staff

July 07, 2023 08:21 Edited

Hi Chiara,

The best way to do this is to simply keep your variables in memory.

model = gp.Model()
x = model.addVars(N)

def update_values(N, x):
    values = []
    for i in range(N):
        value = x[i].X
        values.append(value)
    # also values = [x[i].X for i in range(N)]
    return values

This will give you easy access to the solution values without having to call \(\texttt{model.getVarByName}\).

Cheers,
David

Chiara

July 07, 2023 11:32

Thanks a lot David :)

This is the perfect way to solve my problem.

Have a nice day!

Branch and Bound algorithm

Comments

Didn't find what you were looking for?