san99tiago/ml_basics

Machine Learning theory with Python and MATLAB | Development of robust algorithms from scratch.

👶 🤖 MACHINE LEARNING BASICS 🤖 👶

This repository covers the basic concepts of Machine Learning.
It begins with the necessary programming tools to understand the fundamental knowledge for development of ML solutions.
Then, it goes into more advanced topics and problem-solving with powerful ML approaches.
The information of these repository is based on some lectures and my own presonal projects:

• MIT 6.034: Artificial Intelligence (Fall 2010)
  https://www.youtube.com/playlist?list=PLUl4u3cNGP63gFHB6xb-kVBiQHYe_4hSi

• Stanford CS229: Machine Learning (Autumn 2018)
  https://www.youtube.com/playlist?list=PLoROMvodv4rMiGQp3WXShtMGgzqpfVfbU

• Udemy course: Python for Data Science and Machine Learning Bootcamp
  https://www.udemy.com/course/python-for-data-science-and-machine-learning-bootcamp/#instructor-1

• Lectures of "Intelligent Systems" from the EIA University.

Dependencies 🚦

The dependencies are mainly python libraries and the correct usage of them.
My advice is to primary understand the "pip" package installer, before working with other dependencies.

Software dependencies

• VS Code
  Visual Studio Code is the main code editor for Python scripts and high-level programming. This is not absolutely necessary, but from my experience, it gives us a great performance and we can link it with Git and Github easily.

https://code.visualstudio.com/

• MATLAB
  MATLAB is a great tool to work with Machine Learning Algorithms, however, you require a license to work with it. My goal is to make a transition from MATLAB to Python in the long term. Simulink is a great tool for Dynamic Systems.

https://www.mathworks.com/products/matlab.html

• Python
  Python is the main programming language that let us work fast, with easy and powerful integration of different software solutions.

https://www.python.org/

Libraries and Package dependencies

• NUMPY
  NumPy is a great resource to work with arrays and concepts of linear algebra. This allows us to work with data and achieve scientific computing

https://numpy.org/

• PANDAS
  This is a powerful tool to manipulate and work with data. It is simple to use and built based on Python and NumPy library.

https://pandas.pydata.org/

• MATPLOTLIB
  This is a great Python library for creating static, animated and interactive data visualizations in a simple way.

https://matplotlib.org/

• SEABORN
  Library built "on top" of matplotlib, to create simple and powerful statistics plots.

https://seaborn.pydata.org/

• TKINTER
  Tkinter is an integrated Library that gives us a great way to create GUI windows with Python. It is also a cross-platform that has the advantage of working with the classic Tk windows and many libraries that integrate together to achieve amazing projects.

https://docs.python.org/3/library/tkinter.html

• PYINSTALLER
  Pyinstaller is a simple and effective way to generate executable files based on the applications developed in python language. It gives us a lot of variations, and production options. It's important to look for its documentation for the correct usage.

https://www.pyinstaller.org/

Usage ✏️

All projects are really well commented and most of them have specifications and remarks for their purpose.

I will be uploading most of the files, and try to keep it as clean as possible.

Special thanks 🎁

• Thanks to the MIT professor of an online course: Patrick Winston.
• Thanks to the Stanford professor of an online course: Andrew Ng.
• Thanks to the Udemy professor of an online course: Jose Marcial Portilla.
• Thanks to all contributors for the great OpenSource projects that I am using.

Author 🎹

Santiago García Arango.