in Python – shared memory space ## How? Python/C API Cython ○ Numba / PyPy Boost::Python ○ Pybind11 ○ cppyy ☐ Mix and match! ## Hello, world (not really) ## HELLO, WORLD Different implementations increment function (takes an integer as argument, increments it and returns) C++ • Python • C++ with pybind11 C++ with boost::python • Cython int increment(int); // defined in a separate translation unit 09a979c32/p19_1.jpg) Cython PyPy New "C++" code Focus of this talk Boost::Python Pybind11 Cython Cython cppyy ## Performance ## SOME PERF NUMBERS: PER INCREMENT RUNTIME Take these numbers

f/dcff3d9f70403fc7fe9fcc283f05b024/p2_2.jpg) ## 大家以前都是怎样调用C++代码呢？ 1. Cython 2. boost.python/pybind11 3. ctypes  ## 先不讨论这些工具的优劣 b024/p6_2.jpg) ## 与其他工具的用法进行一下比较？ ## 与pybind11进行比较 • c2py: 两个命令  • pybind11：1137行代码？   ## 跟pybind11比算个什么事？ 不在一个次元！ ## 要比就和不需要写C++代码的工具比！ ## 与ctypes进行比较 • c2py:调用代码，并享受type hint。 • ctypes:调用代码？

interesting thing is that nothing has to be done in advance for that library, thanks to useful tools as pybind11, that allows to create python bindings easily. So let’s build a package with the following files: The package recipe. • __init__.py: necessary file, blank. • pypoco.cpp: The C++ code with the pybind11 wrapper for Poco that generates a python extension (a shared library that can be imported from python) thanks to the elegant pybind11 library: #include <pybind11 pybind11.h=""> #include "Poco/Random.h" using Poco::Random; Listing 8: pypoco.cpp (continues on next page)pybind11> (continued from previous

interesting thing is that nothing has to be done in advance for that library, thanks to useful tools as pybind11, that allows to create python bindings easily. So let’s build a package with the following files: py: The package recipe. • `init_.py: necessary file, blank.` • pypoco.cpp: The C++ code with the pybind11 wrapper for Poco that generates a python extension (a shared library that can be imported from python) thanks to the elegant pybind11 library: #include <pybind11 pybind11.h=""> #include "Poco/Random.h" using Poco::Random; Listing 8: pypoco.cpp (continues on next page)pybind11> (continued from previous

interesting thing is that nothing has to be done in advance for that library, thanks to useful tools as pybind11, that allows to create python bindings easily. So let’s build a package with the following files: The package recipe. • `__init__.py`: necessary file, blank. • pypoco.cpp: The C++ code with the pybind11 wrapper for Poco that generates a python extension (a shared library that can be imported from python) easily thanks to the elegant pybind11 library: Listing 10: pypoco.cpp #include <pybind11 pybind11.h=""> #include "Poco/Random.h" using Poco::Random; namespace py = pybind11; PYBIND11_PLUGIN(pypoco)

interesting thing is that nothing has to be done in advance for that library, thanks to useful tools as pybind11, that allows to create python bindings easily. So let’s build a package with the following files: The package recipe. • __init__.py: necessary file, blank. • pypoco.cpp: The C++ code with the pybind11 wrapper for Poco that generates a python extension (a shared library that can be imported from python) easily thanks to the elegant pybind11 library: Listing 10: pypoco.cpp #include <pybind11 pybind11.h=""> #include "Poco/Random.h" using Poco::Random; namespace py = pybind11; PYBIND11_PLUGIN(pypoco)

thing is that nothing has to be done in advance for that library, thanks to useful tools such as pybind11, that lets you easily create Python bindings. So let’s build a package with the following files: • __init__.py: A required file which should remain blank. • pypoco.cpp: The C++ code with the pybind11 wrapper for Poco that generates a Python extension (a shared library that can be imported from Python) easily thanks to the elegant pybind11 library: Listing 10: pypoco.cpp #include <pybind11 pybind11.h=""> #include "Poco/Random.h" using Poco::Random; namespace py = pybind11; PYBIND11_PLUGIN(pypoco)

thing is that nothing has to be done in advance for that library, thanks to useful tools such as pybind11, that lets you easily create Python bindings. So let’s build a package with the following files: • __init__.py: A required file which should remain blank. • pypoco.cpp: The C++ code with the pybind11 wrapper for Poco that generates a Python extension (a shared library that can be imported from Python) easily thanks to the elegant pybind11 library: Listing 10: pypoco.cpp #include <pybind11 pybind11.h=""> #include "Poco/Random.h" using Poco::Random; namespace py = pybind11; PYBIND11_PLUGIN(pypoco)

1/6/5/71656c39f0055537d7f9feafcf0f03f1/p28_2.jpg) ## pybind c++ extension • pybind11 • 感谢知乎cuBERT提供的c++实现 • 用pybind11一波封装 • 再加上正经多线程 ![Image](/uploads/documents/7/1/6/5/71656c39f0055537d7f9feafcf0f03f1/p29_2

thing is that nothing has to be done in advance for that library, thanks to useful tools such as pybind11, that lets you easily create Python bindings. So let's build a package with the following files: - __init__.py: A required file which should remain blank. - pypoco.cpp: The C++ code with the pybind11 wrapper for Poco that generates a Python extension (a shared library that can be imported from Python) thanks to the elegant pybind11 library: Listing 22: pypoco.cpp ```cpp #include <pybind11 pybind11.h=""> #include "Poco/Random.h" using Poco::Random; namespace py = pybind11; PYBIND11_PLUGIN(pypoco)