– explained by a data scientist
When you look all over the internet, there are a lot of articles telling you that you can learn data science in 24-hours. Others have even gone to an extent of telling you that it is possible to learn everything from python ‘hello world’ to deploying a machine learning model in just three hours. Well…is that true?
Look…I will be honest with you. Learning data science from scratch can take you years and years. That is true if you do not know what you are doing. By the end of this article, I am going to tell you what exactly works. I will tell you what worked for me and a few other outlier data scientists who know what they are doing.
From my experience, the process of learning data science effectively can be narrowed down to 3 months. And here is my breakdown on how you can achieve your dream of becoming a data scientist in 3 months:
1st Month: Choose and Learn a Good Data Analytics Tool/Programming Language
There are many data science tools out there. Mentioning all of them will leave you confused. The truth is you do not have to spend 10 years searching for the best programming language for data science. Because if you do, you would be old before you get started. I will recommend two programming languages for data science – Python and R. Should you learn both of them at the same time? NO. Start learning one of the languages (don’t overthink on which to choose). For my case, I have knowledge in both languages. However, I recommend Python because it is versatile, easy to learn and good for the job (that’s my bias recommendation).
No matter what language you will choose, you will get to the same destination. Learn the language from its basics to its advanced concepts. Like I said in this article, the skeleton (concepts) of all programming languages is the same. Read the article to get guide on how to learn any programming language effectively.
2nd Month: Learn How to Do Data Analysis Using Your Chosen Language/Tool
Python has its libraries and packages designed for data analysis. So does R and other tools. Examples of Python’s data science packages are numpy, pandas, seaborn, matplotlib, scipy etc. Examples in R include tidyverse mega package (a collection of data analysis and visualization packages). Learn the data science packages for your chosen language slowly but effectively. Do exercises (a.k.a milestone projects) after learning each package. This will help you understand each package/library and its importance in the large data science tasks.
3rd Month: Do at Least 3 Big Data Science Projects
You can call them milestone projects. There are lots of resources on the web to help you find beginner Data Science projects to start with. My best go-to platform for projects is Kaggle. Kaggle is an online platform where data scientists can get projects and competitions to practice and perfect their skills.
There are two main reasons why I recommend doing at least 3 data science projects:
- Projects will take you from the learning stage to the tackling real-world problems stage. You do not know how good you are at something until you face a challenge. Through tackling and solving different projects, you are able to tell what areas or concepts you have mastered – and what you need to revisit. Plus, you will get to learn how to think outside the box when facing new and different data science challenges even in future. That is how you can assess yourself. And it is how you get to grow.
- Projects are good for your portfolio. That is what you are going to show to your potential employers out there if you ever look for a job. You can also share with other learners in case you want to share your knowledge. Even when you will be looking for potential project partners to work with, that will be your selling point. You will have something to showcase your skills and abilities.
And that is my 3-months strategy to learning data science. If you stay committed, dedicate your time and follow the above strategy…you will see results.
