Hey, welcome on bord, here you can follow my progress in data science. Below I have mapped the learning path
I am following. I will update that path as I discover more content. Of course, it is not an exhaustive list.
In the next following sections, we will go through all this different fields.
First, it is important to have a good overview of this broad subject. In that purpose, understanding the job of a data scientist is primordial. As a resume, I gather in the list and the mind map bellow all the “tools” you and I will need to master in Data Science:
As a spark of motivation I start by understanding the basic of Machine Learning. The goal here is to gain sens and motivation for the futur work in math and code. I recommend Vishal Maini & Samer Sabri’s book, this is a good resource to start with the big picture of artificial intelligence and machine learning.
This series is a guide for getting up-to-speed on high-level machine learning concepts. It is easy to understand with a good mixte between analogies, examples and technical explanation. You will find a lot of different resources to go deeper in the different fields - projects, videos, books and so much more. The authors give as well some interesting personal advice at the end of the book.
What I have learned?
To have a deeper understanding of machine learning, I review some mathematical knowledge. My goal is about creating mathematical intuition rather than learn the formulas and computation methods.
Essence of Linear Algebra & Essence of Calculus: I highly recommend to see the videos of Grant Sanderson. His visual approach with superb animations will give you a solid conceptualization of those subjects.
MVC Khan Academy: Khan Academy and Grant Sanderson worked together in the purpose to create a course with a visual meaning of important tools useful in the field of Multivariable Calculus. I realy enjoy the part on Maxima and Minima and Constrained Optimization. You will learn about Gradient, Laplacian, Lagrangian multiplier and so on.
Statistics and probability Khan Academy: This interactive and well-designed course provides an introduction to important tools as Regression, Confidence Intervals, Significance Test and so much more.
What I have learned?
If you want to get into Data Science, you have to know how to manipulate data. Most of your time, as a data scientist engineer, will be spend on manipulating, analysing and getting data ready for modeling. I recommend to perform data analysis with pandas, the popular open source library for data analysis in Python.
One of the creator of pandas, Wes McKinney, wrote a book called “Python Data Analysis” and it is a really good choice for learning data analysis in Python. This book show you how to solve a broad set of data analysis problems effectively. You will get complete instructions for manipulating, processing, cleaning, and crunching dataset. It is well written and the real-world examples are really usefull to understand the workflow of data analysis in Python.
As a complement, in the Harvard course on Data science, you will find great lectures on data analysis. I really recommend to at least take a look to Exploratory Data Analysis and effective Visualizations, you will discover some usefull principles for effective visualization. Take also a look at plotly and Tableau to complete your skills in data visualization.
What I have learned?
Through the book Hands‑On Machine Learning with Scikit‑Learn, Keras, and TensorFlow by Aurélien Géron I discovered the machine learning landscape. Each chapter ends by a quizz and few coding project. A really enjoy this mix between learning new knowledge, testing your understanding with the quizz and practising on project.
What I have learned?
“Value of knowledge shouldn’t depend upon where that knowledge was acquired” — Giles McMullen-Klein. Special thanks to this guy who gathers a lot of different resources about data science. Those videos are superb and motivational.