CODEWITHSUNDEEP
pythonlinear-programmingor-toolsobjective-function
JGW
JGW

Reputation: 334

How can I correct objective function error in OR Tools in Python?

I'm working on a variation of the 0-1 knapsack problem using OR tools. I've managed to solve one type of variation, however, on this attempt, I'm having difficulty specifically with the objective function. Here is the code that I'm using. Basically, I want to sum each of the variables associated with an item in the objective function, and then if the item is selected the x[i,j] binary variable will be set to 1:

import math
import pandas as pd
import ortools
from ortools.linear_solver import pywraplp

solver = solver = pywraplp.Solver.CreateSolver('SCIP')

#All of the variables in the dataframe are integer variables
df = pd.read_csv('data.csv') 

#Create a data dictionary to store all of the data items in
data = {}

data['ID'] = df.ID 
data['time'] = df.Time
data['days'] = df.Days
data['canc'] = df.Canc
data['hdu'] = df.HDU
data['priority'] = df['Priority']
  
data['items'] = list(range(len(df.ID)))
data['num_items'] = len(df.ID)
num_knapsacks = 2
data['capacities'] = [480,240]

data['knapsacks'] = list(range(num_knapsacks))

x = {}
for i in data['items']:
    for j in data['knapsacks']:
        x[(i,j)] = solver.IntVar(0,1,'x_%i_%i' % (i, j))
        
#Constraint to limit item to a single knapsack
for i in data['items']:
    solver.Add(sum(x[i,j] for j in data['knapsacks'])<=1)

#Knapsack Capacity Constraint
for j in data['knapsacks']:
    solver.Add(sum(x[(i,j)]*data['time'][i] 
                  for i in data['items']) <= data['capacities'][j])
    
# HDU Constraint
for j in data['knapsacks']:
    solver.Add(sum(x[(i,j)]*data['hdu'][i]
                  for i in data['items']) <= 1)

#objective function
objective = solver.Objective()
for i in data['items']:
    for j in data['knapsacks']:
        objective.SetCoefficient(x[(i,j)], data['days'][i]+data['time'][i]+data['priority'][i]+data['canc'][i])                       
objective.SetMaximization()

The error i'm getting is specifically with the objective function, but I'm not sure why (I got a different version working using a single objective coefficient). The error is:

   return _pywraplp.Objective_SetCoefficient(self, var, coeff)

TypeError: in method 'Objective_SetCoefficient', argument 3 of type 'double'

Upvotes: 1

Views: 641

Answers (1)

Laurent Perron
Laurent Perron

Reputation: 11014

most likely you data is a numpy.float(). Cast it to float in SetCoefficient.

Upvotes: 2

Related Questions