Machine Learning Maverick – My Path of Learning

Preface

In every task I like to see the big picture, the same is true with Machine Learning. When I started to learn ML, I wondered what are the core steps to work in Machine Learning. Reading many articles on the internet, I found some patterns, which create necessary steps while working with Machine Learning.

The Strategy – Big picture while tackling Machine Learning task

1. Data Collection.
2. Data Modeling.
3. Implementation/Deployment.

The Tactics – Necessary steps while working with Machine Learning task

1. Defining the problem
   • What problem are we trying to solve?
2. The data
   • What kind of data do we have?
3. Assessment
   • What is the definition of success?
4. Characteristics
   • What kind of features should we model?
5. Modeling
   • What kind of model should we use?
6. Examination
   • What is the outcome, what more can be done?

When I started to work with Machine Learning I didn’t even realize that so many steps are involved. Each of the steps, strategic or tactic requires separate articles to describe them.

The Toolkit

For each tactical step, a different set of tools is required as well. Below is a list of tools that currently I use during my work on Machine Learning tasks. I will just name them, each tool requires separate articles to describe them.

1. Python.
2. Conda.
3. Jupyter Notebook.
4. pandas.
5. NumPy.
6. matplotlib.
7. scikit-learn.

NOTE: In this article, I’m just barely scratching the surface. This topic needs more reading and research on your own. As a good starting point, readers may check a few useful resources in my other article – How did my journey with AI & ML start?. I’m still at the beginning of my learning process with AI & ML!