PyOptSparse: install others optimizers algorithms

Viewed 32

I am opening this discussion in the hope that I could find some help about how to install the optimizers which are not included within the PyOptSparse library. Indeed, I found that "Some optimizers are proprietary, and their sources are not distributed with pyOptSparse. To use them, please follow the instructions on specific optimizer pages" on the official PyOptSparse website and thus, I would like to install those which are not included such as NOMAD or ParOpt.

https://mdolab-pyoptsparse.readthedocs-hosted.com/en/latest/install.html#install

Have you already tried to do this? Thanks in advance for your help.

1 Answers

You don't have to use the OpenMDAO driver interface to integrate with optimizers if a driver for your needs doesn't exist. There are some nice things about the OpenMDAO driver interface, like the ability to add/remove design variables simply or to change the scaling quickly.

However, you can use any optimizer with its native interface pretty simply. You can follow the no_driver solution approach from OpenMDAO RevHack 2020 for guidance on how to do that.

Related