Sign Up to our social questions and Answers Engine to ask questions, answer people’s questions, and connect with other people.
Login to our social questions & Answers Engine to ask questions answer people’s questions & connect with other people.
Lost your password? Please enter your email address. You will receive a link and will create a new password via email.
Please briefly explain why you feel this question should be reported.
Please briefly explain why you feel this answer should be reported.
Please briefly explain why you feel this user should be reported.
I’m looking for a software package to solve a very large, sparse non-linear least squares problem in C++. I’ve come across a large number of modern linalg libraries in C++ (eigen, armadillo, boost, etc.), but none seem to have such a solver (or even a regular least-squares solver) built in. I’d really like to avoid a bunch of messy calls to an old C / Fortran interface if possible. Thanks!
I would simply use the general-purpose NLP solver IPOPT written in C++. It is the most robust solver among those I have tried and it is meant and successfully used on very large problems.
A change in requirements (e.g. adding constraints) would be no problem if you use the general-purpose IPOPT.
The time consuming part of the solution procedure is to solve the linear systems in each iteration step so it’s worth getting the best linear solver + LinAlg package for your platform.
Unfortunately IPOPT calls Fortran subroutines internally so you will need a Fortran compiler which is sort of a pain.
If IPOPT is not enough, you will have to look for a problem specific solver.