autoware_pyplot

This package provides C++ interface for the notable matplotlib using pybind11 backend for

• creating scientific plots and images illustrating the function inputs/outputs
• debugging the output and internal data of a function before unit testing in a more lightweight manner than planning_simulator

usage

In your main function, setup the python context and import matplotlib

#include <autoware/pyplot/pyplot.hpp>
#include <pybind11/embed.h>

// in main...
  py::scoped_interpreter guard{};
  auto plt = autoware::pyplot::import();

Then you can use major functionalities of matplotlib almost in the same way as native python code.

{
    plt.plot(Args(std::vector<int>({1, 3, 2, 4})), Kwargs("color"_a = "blue", "linewidth"_a = 1.0));
    plt.xlabel(Args("x-title"));
    plt.ylabel(Args("y-title"));
    plt.title(Args("title"));
    plt.xlim(Args(0, 5));
    plt.ylim(Args(0, 5));
    plt.grid(Args(true));
    plt.savefig(Args("test_single_plot.png"));
}

{
    auto [fig, axes] = plt.subplots(1, 2);
    auto & ax1 = axes[0];
    auto & ax2 = axes[1];

    ax1.set_aspect(Args("equal"));
    ax2.set_aspect(Args("equal"));
}

Changelog for package autoware_pyplot

1.0.0 (2025-03-31)

• feat(trajectory): remove default ctor and collect default setting in Builder (#287)
• Contributors: Mamoru Sobue

0.3.0 (2025-03-21)

• chore: fix versions in package.xml
• feat(trajectory): improve comment, use autoware_pyplot for examples (#282) Co-authored-by: Yutaka Kondo <<yutaka.kondo@youtalk.jp>>
• feat: port autoware_pyplot from Autoware Universe to Autoware Core (#199)
• Contributors: Mamoru Sobue, Ryohsuke Mitsudome, mitsudome-r