Quick start

After following the installation guide, you will be able to compile and run your first C++ pagmo program:

getting_started.cpp

#include <iostream>

#include <pagmo/algorithm.hpp>
#include <pagmo/algorithms/sade.hpp>
#include <pagmo/archipelago.hpp>
#include <pagmo/problem.hpp>
#include <pagmo/problems/schwefel.hpp>

using namespace pagmo;

int main()
{
    // 1 - Instantiate a pagmo problem constructing it from a UDP
    // (i.e., a user-defined problem, in this case the 30-dimensional
    // generalised Schwefel test function).
    problem prob{schwefel(30)};

    // 2 - Instantiate a pagmo algorithm (self-adaptive differential
    // evolution, 100 generations).
    algorithm algo{sade(100)};

    // 3 - Instantiate an archipelago with 16 islands having each 20 individuals.
    archipelago archi{16u, algo, prob, 20u};

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

    // 5 - Wait for the evolutions to finish.
    archi.wait_check();

    // 6 - Print the fitness of the best solution in each island.
    for (const auto &isl : archi) {
        std::cout << isl.get_population().champion_f()[0] << '\n';
    }
}

Place it into a getting_started.cpp text file and compile it (for example) with:

$ g++ -O2 -DNDEBUG -std=c++17 getting_started.cpp -pthread -lpagmo -lboost_serialization -ltbb

If you installed pagmo in a non-standard path, such as the .local directory in your $HOME on a Unix installation (e.g., /home/username/.local), the compiler will need assistance to locate the pagmo headers and libraries. E.g., you may need a command such as:

$ g++ -O2 -DNDEBUG -std=c++17 getting_started.cpp -pthread -lpagmo -lboost_serialization -ltbb -I /home/username/.local/include -L /home/username/.local/lib -Wl,-R/home/username/.local/lib

If you installed pagmo with support for optional 3rd party libraries, you might need to add additional switches to the command-line invocation of the compiler.

We recommend to use pagmo’s CMake support in order to simplify the build process of code depending on pagmo (see next section).

Using pagmo with CMake

As a part of the pagmo installation, a group of CMake files is installed into CMAKE_INSTALL_PREFIX/lib/cmake/pagmo. This bundle, which is known in the CMake lingo as a config-file package, facilitates the detection and use of pagmo from other CMake-based projects. pagmo’s config-file package, once loaded, provides an imported target called Pagmo::pagmo which encapsulates all the information necessary to use pagmo. That is, linking to Pagmo::pagmo ensures that pagmo’s include directories are added to the include path of the compiler, and that the libraries on which pagmo depends are brought into the link chain.

For instance, a CMakeLists.txt file for the simple getting_started.cpp program presented earlier may look like this:

# The name of our project.
project(sample_project)

# Look for an installation of pagmo in the system.
find_package(Pagmo REQUIRED)

# Create an executable, and link it to the Pagmo::pagmo imported target.
# This ensures that, in the compilation of 'getting_started', pagmo's include
# dirs are added to the include path of the compiler and that pagmo's
# dependencies are transitively linked to 'getting_started'.
add_executable(getting_started getting_started.cpp)
target_link_libraries(getting_started Pagmo::pagmo)

Place this CMakeLists.txt and the getting_started.cpp files in the same directory, and create a build subdirectory. From the build subdirectory, execute these commands to produce the getting_started executable:

$ cmake ../ -DCMAKE_BUILD_TYPE=Release
$ cmake  --build .

Please refer to the CMake documentation for more information on how to use CMake.