climt VS prima

The developer of PRIMA here.

If you use method "cobyla" from scipy.optimize.minimize, then PRIMA already performs far better (in terms of the number of function evaluations). See the comparison at https://github.com/libprima/prima#improvements .

The bugs are indeed only a secondary reason: they can only be triggered under special situations. They may not affect your usage at all (when it does affect you, the consequence is catastrophophic).

1. nagfor Release 7.1(Hanzomon) Build 7149 released on March 5, 2024, fixed all the bugs spotted, but introduced an ICE when compiling PRIMA ( http://www.libprima.net ). The ICE has nothing to do with half-precision real, because it occurs when PRIMA is configured to use single or double precision. It can be reproduced by

```

git clone https://github.com/libprima/prima.git && cd prima && git checkout ec42cb0 && cd fortran/examples/lincoa && make ntest

```

2. nagfor 7.2 released on 6 March, 2024 included neither the ICE nor the fixes for the bugs.

My goal is to rewrite this library for derivative-free optimization: https://github.com/libprima/prima

A native port is indeed planned. However, since we are talking about a project of about 10K lines of code, such a port will not be delivered very soon.

In fact, native implementations of PRIMA in Python, MATLAB, C++, Julia, and R will all be done in the future. See https://github.com/libprima/prima#other-languages . But it takes time. PRIMA has been a one-man project since it started three yearss ago. Community help is greatly needed.

Thanks.

GitHub repo of the project: https://github.com/libprima/prima

It sounds like this was a difficult task. The motivation to fulfill Prof. Powell's request and help the community of derivative-free optimization users must have been strong. Congratulations on your achievement!

From the GitHub README:

> In the past years, while working on PRIMA, I have spotted a dozen of bugs in reputable Fortran compilers and two bugs in MATLAB. Each of them represents days of bitter debugging, which finally led to the conclusion that it was not a problem in my code but a flaw in the Fortran compilers or in MATLAB. From a very unusual angle, this reflects how intensive the coding has been.

> The bitterness behind this "fun" fact is exactly why I work on PRIMA: I hope that all the frustrations that I have experienced will not happen to any user of Powell's methods anymore. I hope I am the last one in the world to decode a maze of 244 GOTOs in 7939 lines of Fortran 77 code — I have been doing this for three years and I do not want anyone else to do it again.

https://github.com/libprima/prima#a-fun-fact