Quick start

C++

After following the Installation you will be able to compile and run your first C++ pagmo_plugins_nonfree program. Let us have a look at a mininal cpp file and its corresponding CMakeLists.txt:

#include <pagmo/algorithm.hpp>
#include <pagmo/archipelago.hpp>
#include <pagmo/problem.hpp>
#include <pagmo/problems/rosenbrock.hpp>
#include <pagmo_plugins_nonfree/pagmo_plugins_nonfree.hpp>

using namespace pagmo;

int main()
{
    // 1 - Instantiate a pagmo problem constructing it from the Rosenbrock
    // UDP (user defined problem).
    problem prob{rosenbrock(30)};

    // 2 - Instantiate a pagmo_plugins_nonfree algorithm, in this case SNOPT. This assumes a library libsnopt7_c.so is
    // found in the path "/usr/local/lib/". Otherwise you will get a runtime error.
    algorithm algo(snopt7(false, "/usr/local/lib/libsnopt7_c.so"));

    // 3 - Instantiate an archipelago with 16 islands having each 1 individual (the initial guess)
    archipelago archi{16, algo, prob, 1};

    // 4 - Run the snopt7 calls in parallel on the 16 separate islands.
    archi.evolve(1);

    // 5 - Wait for the snopt7 calls to be finished
    archi.wait();

    // 6 - Print the fitness of the best solution in each island
    for (const auto &isl : archi) {
        print(isl.get_population().champion_f(), "\n");
    }
}

cmake_minimum_required(VERSION 3.20)
project(getting_started_example LANGUAGES CXX)
set(CMAKE_CXX_STANDARD 17)
set(CMAKE_CXX_STANDARD_REQUIRED ON)

# Find pagmo and the plugins package installed by this project.
find_package(pagmo REQUIRED)
find_package(pagmo_plugins_nonfree REQUIRED)

add_executable(getting_started getting_started.cpp)

# Link the pagmo_plugins_nonfree exported target and pagmo
target_link_libraries(getting_started PRIVATE Pagmo_plugins_nonfree::pagmo_plugins_nonfree Pagmo::pagmo)

Make a folder and place there the C++ code into a getting_started.cpp text file and the CMake code into a CMakeLists.txt text file. Then, from the command line, you can build and run the example with:

$ mkdir build && cd build
$ cmake ..
$ cmake --build .

---

Python

If you have successfully installed pygmo_plugins_non_free following the Installation you can try the following script:

import pygmo as pg
import pygmo_plugins_nonfree as ppn

# 1 - Instantiate a pygmo problem constructing it from a UDP
# (user defined problem).
prob = pg.problem(pg.schwefel(30))

# 2 - Instantiate a pagmo_plugins_nonfree algorithm, in this case SNOPT.
# Here we assume the library name is libsnopt7_c.so, in Windows it would probably be
# snopt7_c.dll or similar.
algo = pg.algorithm(ppn.snopt7(false, "/usr/local/lib/libsnopt7_c.so"))

# 3 - Instantiate an archipelago with 16 islands having each 20 individuals
archi = pg.archipelago(16, algo=algo, prob=prob, pop_size=20)

# 4 - Run the evolution in parallel on the 16 separate islands 10 times.
archi.evolve(10)

# 5 - Wait for the evolutions to be finished
archi.wait()

# 6 - Print the fitness of the best solution in each island
res = [isl.get_population().champion_f for isl in archi]
print(res)

Place it into a getting_started.py text file and run it with:

python getting_started.py

We recommend the use of Jupyter or Ipython to enjoy pygmo and its affiliated packages the most.