Python and Machine Learning: From 0 to 100 with Reinforcement Learning

About the Course

On March 24, 2020, I had the pleasure of teaching a course titled “Python and Machine Learning: From 0 to 100 with Reinforcement Learning” in collaboration with MalagaAI. This course was designed as an online session to teach participants how to train Reinforcement Learning models using Python.

The course was organized by MalagaAI, a community dedicated to promoting artificial intelligence in Málaga, with the collaboration of Andrés Matesanz and Joaquín Terrasa.

Chapter 1: A Python Padawan

The main course content is available in the GitHub repository. Chapter 1, titled “A Python Padawan”, serves as a fundamental introduction to Python programming oriented towards Machine Learning.

Chapter Content

This chapter covers three main blocks:

1. Better and simpler: how to be a zen master (of Python)

   • Python programming practices
   • Creating reproducible code
   • Python versions (from 2 to 3.x)
   • Help functions and reserved words
   • Python standard library

2. Functional Programming

   • Fundamental concepts of functional programming
   • Applications in data science

3. Numerical Computing

   • Tools for efficient numerical calculations
   • Preparation for Machine Learning

Why We Use Python

The course explains why Python is the preferred language for Machine Learning:

• Readability: Python uses English words and a fair balance between productivity and maintenance
• Multiparadigm: Allows programming using object-oriented and functional models
• Dynamic typing: Doesn’t require explicit type assignment, although optional static typing is allowed
• Robust standard library: With over 211 modules in version 3.7
• Great community: Open source with an active community that proposes improvements
• Portability: Present on multiple platforms including Linux and macOS

Duration and Format

• Estimated duration: 3 hours (2 in streaming + 1 individual work)
• Format: Online session with practical demonstrations

How to Access the Material

The course content is available in several ways:

1. GitHub: Clone the repository by downloading the .zip or using git clone
2. Binder: Free cloud notebook environment available at mybinder.org
3. Google Colab: Interactive version available at Google Colab

Related Resources

• Previous chapter: Introduction to Python by Andrés Matesanz
• Official Python documentation: python.org
• Language rankings: PYPL and TIOBE

Note for Students

This course is designed to be reproducible and accessible. If you have suggestions or find errors, please share them in the repository to continue improving the material.

I hope you enjoy learning Python and Reinforcement Learning!