How do I consume numpy arrays in C++ with pybind11?

Question

Even after reading through the Pybind11 documentation on numpy, I'm still unsure how to consume a numpy array in C++, for example by converting it into a vector if it's 1-dimensional, or a vector of vectors if it's two dimensional.

Here's a snippet of my code. How would I go about implementing make_vector_from_1d_numpy_array()?

#include <pybind11/embed.h>
#include <pybind11/numpy.h>
namespace py = pybind11;
using namespace pybind11::literals; 

py::scoped_interpreter guard {};
py::module np = py::module::import("numpy");
py::module librosa = py::module::import("librosa");

auto filters_py = librosa.attr("filters").attr("mel")(16000, 1024, "n_mels"_a = 80, "fmin"_a = 0, "fmax"_a = 8000, "htk"_a = true);
auto shape_py = filters_py.attr("shape");

// above code runs fine. At this point, shape_py is "numpy.ndarray"
// auto shape = make_vector_from_1d_numpy_array<size_t>(shape_py)

Answer 1

This works.

template<class T>
std::vector<T>make_vector_from_1d_numpy_array( py::array_t<T>py_array )
{
    return std::vector<T>(py_array.data(), py_array.data() + py_array.size());
}