-2

CVXPY is a package that enables convex optimization, but the community version of CPLEX only supports 1000 variables and constraints (https://www.cvxpy.org/api_reference/cvxpy.problems.html#id3).

I need to optimize a portfolio larger than 1000, but I can't seem to find information on how to on here: https://www.cvxpy.org/install/

Thanks!

Aj LaPanta
  • 11
  • 1
  • 5
  • 2
    This has nothing to do with cvxpy which is a pure free open source project (focusing on modelling and transformations; not solvers). It has multiple solver backends, one of those being the commercial software cplex (not related to cvxpy). Visit [their homepage](https://www.ibm.com/analytics/cplex-optimizer). cvxpy itself also links to [external pages](https://www.cvxpy.org/install/#install-with-cplex-support) in regards to cplex. (Before wasting time: If you are doing academic work, there might be acad. licenses, if your company pays: fine; obtaining a license as a private person is rare imho) – sascha Dec 08 '21 at 00:51

1 Answers1

0

CPLEX is only one of many CVXPY backends as stated by @sascha in the comment.

If you have academic background (staff, student, teacher, etc), you can request free academic CPLEX via IBM Data Science.

Step by step:

  1. Navigate to Software > ILOG > Download.

  2. Find ILOG that satisfy your need and python version.

  3. Install ILOG

  4. Find the location of cplex setup.py in your ILOG installation and install it using python setup.py install.

See ILOG CPLEX Optimization Studio docs here.

If you are interested with other backend, see the installation guide of CVXPY

Muhammad Yasirroni
  • 1,512
  • 12
  • 22