How import cplex in google colab? - google-colaboratory

Question

!apt install cplex-utils
!pip install cplex
solver = SolverFactory('cplex')
res_NLP= solver.solve(HN_model)
The error is:
WARNING: Could not locate the 'cplex' executable, which is required
for solver
cplex
--------------------------------------------------------------------------- ApplicationError Traceback (most recent call
last) in ()
1 solver = SolverFactory('cplex')
----> 2 res_NLP= solver.solve(HN_model)
2 frames
/usr/local/lib/python3.7/dist-packages/pyomo/opt/solver/shellcmd.py in
available(self, exception_flag)
123 if exception_flag:
124 msg = "No executable found for solver '%s'"
--> 125 raise ApplicationError(msg % self.name)
126 return False
127 return True
ApplicationError: No executable found for solver 'cplex'

Answer 1

Within IBM Watson Studio, CPLEX comes pre-installed in the Notebooks. But with other Notebook cloud providers, you need to find a way to install it, or else call CPLEX as a service in the IBM Cloud.
You could try to use dowml : https://xavier-nodet.medium.com/submit-decision-optimization-jobs-to-wml-using-dowml-be26e0de6b7f
Or directly wml : https://pypi.org/project/ibm-watson-machine-learning/
With google colab
!pip install cplex
!pip install docplex
from docplex.mp.model import Model
mdl = Model(name='buses')
nbbus40 = mdl.integer_var(name='nbBus40')
nbbus30 = mdl.integer_var(name='nbBus30')
mdl.add_constraint(nbbus40*40 + nbbus30*30 >= 300, 'kids')
mdl.minimize(nbbus40*500 + nbbus30*400)
mdl.export("buses.lp")
!cat buses.lp
works fine and gives
Requirement already satisfied: cplex in /usr/local/lib/python3.7/dist-packages (20.1.0.1)
Requirement already satisfied: docplex in /usr/local/lib/python3.7/dist-packages (2.22.213)
Requirement already satisfied: six in /usr/local/lib/python3.7/dist-packages (from docplex) (1.15.0)
\ This file has been generated by DOcplex
\ ENCODING=ISO-8859-1
\Problem name: buses
Minimize
obj: 500 nbBus40 + 400 nbBus30
Subject To
kids: 40 nbBus40 + 30 nbBus30 >= 300
Bounds
Generals
nbBus40 nbBus30
End

Answer 2

From the error message, SolverFactory seems to be a Pyomo class, and require the CPLEX Interactive executable program to be available locally on the machine where the Pyomo code is executed.
Unless you have a way to install arbitrary executable files on the platform you use, which I very highly doubt if you're not using your own computer, you will have to find another way. Alex's answer proposes two...

Answer 3

When you're working in the colab you need to install cplex from pip. When you install cplex from pip, you need to use the cplex_direct interface in pyomo in order to avoid such errors, since the cplex interface will use the shell approach to solve the problem.
Using Google Colab, this should work
!pip install pyomo -q
!pip install cplex -q
import pyomo.environ as pyo
model = pyo.ConcreteModel()
model.s = pyo.Set(initialize=[1,2,3,4,5])
model.x = pyo.Var(model.s, domain=pyo.NonNegativeReals)
model.c = pyo.Constraint(expr=model.x[model.s.last()]>=5)
model.obj = pyo.Objective(expr=sum(model.x[s] for s in model.s), sense=pyo.minimize)
solver = pyo.SolverFactory('cplex_direct')
solver.solve(model)
model.x.display()
x : Size=5, Index=s
Key : Lower : Value : Upper : Fixed : Stale : Domain
1 : 0 : 0.0 : None : False : False : NonNegativeReals
2 : 0 : 0.0 : None : False : False : NonNegativeReals
3 : 0 : 0.0 : None : False : False : NonNegativeReals
4 : 0 : 0.0 : None : False : False : NonNegativeReals
5 : 0 : 5.0 : None : False : False : NonNegativeReals
I don't use CPLEX a lot, therefore, I'm not fully sure, but I think this free approach should have a limit in the number of var, constraints or others