<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Python on Taha Draidia</title><link>https://tahadraidia.com/tags/python/</link><description>Recent content in Python on Taha Draidia</description><generator>Hugo -- gohugo.io</generator><language>en-us</language><lastBuildDate>Sun, 04 Oct 2020 00:00:00 +0000</lastBuildDate><atom:link href="https://tahadraidia.com/tags/python/index.xml" rel="self" type="application/rss+xml"/><item><title>Let's build a Python module in C</title><link>https://tahadraidia.com/posts/python-c-module/</link><pubDate>Sun, 04 Oct 2020 00:00:00 +0000</pubDate><guid>https://tahadraidia.com/posts/python-c-module/</guid><description>Background Python is one of the most used programming language nowadays specially due its popularity in data science, deep and machine learning fields; Truth to be told, under the hood there is C and/or C++ code running.
Let&amp;rsquo;s take for example some popular python libraries used in math and machine learn to illustrate that point:
Numpy (33.2% of the code is written in C) TensorFlow (61% of the code is written in C++) PyTorch (53% of the code in written in C++ and 4% in C) The raison that these libraries are build in C and/or C++ is for performance issues mainly, also it is important to note that Python is written in C, hence it provides a C API to extend the language by creating new modules at lower-level possible.</description></item></channel></rss>