CODEWITHSUNDEEP
pythonwarningspulp
Axel
Axel

Reputation: 2725

Python PuLP "Overwriting previously set objective." and __dummy = None

I have built a pretty complex MIP in Python PuLP. Obviously a bit too complex for me. When I run it it gives the following Warning:

UserWarning: Overwriting previously set objective. warnings.warn("Overwriting previously set objective.")

The problem performs some calculations but does not come to the expected solution.

The LpStatus[prob.status] returns Optimal

The prob.objective returns None

When I print the prob.variables() with 

for v in prob.variables():
    print(v.name, "=", v.varValue)

I get __dummy = None in between the other variables.

Can anyone explain what the __dummy = None means in this context? Where could I look for a solution? I was pretty sure that I have only one objective function.

Upvotes: 3

Views: 4067

Answers (1)

Axel
Axel

Reputation: 2725

Alright, I found the solution. I indeed overwrote the objective function without noticing it.

Here is the piece of code that caused the warning:

for i in range(len(items)):
        for l in range(L):
            prob += delta[0-l] == 0

the delta variable is a list of lists. The first index was missing, the program therefore compared a list with a zero. Since this is not possible, the equation always returns false (maybe coded as zero) which was then interpreted as an objective function.

This solved the issue:

for i in range(len(items)):
        for l in range(L):
            prob += delta[i][0-l] == 0

Now it adds equations instead of values to the prob.

I hope this helps anyone encountering a similar problem.

Upvotes: 10

Related Questions