Question 1

How to install matplotlib-cpp on Windows with Anaconda?

Accepted Answer

Set PYTHONHOME to the Anaconda directory and QT_QPA_PLATFORM_PLUGIN_PATH as per the README's troubleshooting note, then compile with appropriate -I and -l flags for Python libraries.

Question 2

Is matplotlib-cpp thread safe for concurrent plotting?

Accepted Answer

No, the library is not thread-safe due to Python's interpreter design. You must protect all accesses with a mutex, as stated in the README's issues section.

Question 3

matplotlib-cpp vs Gnuplot for C++ plotting?

Accepted Answer

matplotlib-cpp offers a more familiar API and higher-quality output via matplotlib, but Gnuplot is often lighter and doesn't require Python, making it better for minimal-dependency or performance-critical C++ projects.

Question 4

How to save plots without displaying them in matplotlib-cpp?

Accepted Answer

Use plt::save() directly instead of plt::show(), as shown in the basic example where save() writes the image to a file without opening a window.

Question 5

Can I use matplotlib-cpp with Python 3 instead of Python 2?

Accepted Answer

Yes, the C++ code supports both, but the CMakeLists.txt defaults to Python 3; you may need to adjust compilation flags manually, as noted in the installation section.

Question 6

Why does matplotlib-cpp require linking against libpython?

Accepted Answer

It embeds a Python interpreter internally to call matplotlib, so libpython is essential for spawning the interpreter, which limits use in processes with other Python-embedding libraries.

matplotlib-cpp

What is matplotlib-cpp?

Overview

Use Cases

Best For

Related Projects

Found a gem we're missing?

Not Ideal For

Pros & Cons

Pros

Cons

Frequently Asked Questions