What is it like to step into the hype of Data Science?

You all must have heard about Data Science. After all it was called the “sexist job of the 21st century”. Data science sounds a lot fascinating and involves using cool python libraries with which one can create amazing projects like Facial Recognition, Image Detection etc.

Naturally, the second thing that comes in our mind is who can do Data Science? Is it just for Computer Science students or just can anyone step into the humongous field of Data Science? Even I had the same questions and being a tech-savvy person I was drawn in by the charisma of Data Science. In this post, I’ll share the experience of what it’s really like to study Data Science and will even recommend fantastic resources to get you up and running

Anyone can step into Data Science irrespective of this branch. I am an Electrical Engineering student and currently in my final year. The Final year of engineering is tense and busy because one has to manage curriculum, campus placements, and projects. Right off the bat, if you are thinking about Data Science, you will have to go through the off-campus procedure and will have to keep an eye on job openings on platforms like LinkedIn, etc.

I started learning Data Science in December 2020 and has been well over 5 months since I started. Data Science is a domain that requires a ton of skills and I mean it when I say “a ton of skills”. The most important things which you will come across when you will start is choosing a programing language, learning mathematics and Machine Learning Models.

Note: You will not be able to learn everything from one place. You have to use different platforms like YouTube, Udemy, Coursera, Edx to learn different topics. Data Science is a vast topic and not all things are covered in a single course. Learn to explore different platforms and start reading data science related blogs to be up to date about the latest trends.

1. Selecting a Programing Language

There are two majorly used programing languages for Data Science namely Python and R. If you are a beginner or an intermediate level programmer, Python is a good start. I was already quite familiar with Java and the transition to Python was relatively easy. There is a trauma amongst people for learning a new programming language because it means you will have to start from scratch and learn all the way up. Python is a good start but there may be cases where you have to learn R in the future, maybe it’s for a client project or the company requirement so be prepared for that. For the time being, Python is an excellent option and has a ton of libraries that makes your Data Science journey easier. In Python, there are 3 libraries that you will be using the most: Numpy, Pandas, and Matplotlib. Of these three, Pandas and Matplotlib are extremely crucial. Pandas is used for dataset manipulation and Matplotlib for data visualization. You will have to dedicate time to learn these two libraries for sure. I would highly recommend learning the basics of these two libraries from a YouTube channel Corey Schafer which you can check out here.

2. Mathematics

Statistics and Probability are the two topics that you will be used in Machine Learning. Although there are other topics like Matrics, Calculus etc that are required for Deep Learning and Neural Networks, those will be used later and I would suggest learning the topics as the situation arises. There is no need to learn all the topics at once since it would take a lot of time. From my personal experience, start with Statistics and Probability for Machine Learning, I would highly suggest a Udemy course: The Data Science Course 2021: Complete Data Science Bootcamp. This is an amazing course that consists of Mathematics for ML, some ML models, Deep Learning and Neural Networks. The Mathematics part is explained beautifully with animations. You can easily get the course for like ₹400 . This course doesn’t cover every single thing but instead is aimed at beginners. I myself found this course very helpful.

3. Machine Learning Models

There are many ML Models to learn. Some popular models are Linear Regression, Logistic Regression, K Means, etc. The most important part in learning an ML model is understanding the mathematics part and the intuition being the model as to how exactly it works. You may think why would one need to waste time to understand how the model works since it’s already doing the job just right. The answer to this question is - for Optimization. You have to understand how the model works exactly to perform optimizations. The syntax is pretty much easy to learn and implement. I would highly recommend a YouTube channel Krish Naik for understanding the underlying intuition of models. He is an extremely humble person who is contributing to the data science community and you won’t regret watching his videos. If you want a cheap course, I would again recommend a Udemy course: Machine Learning A-Z™: Hands-On Python & R In Data Science. The reason why I am recommending this course is that this course has almost all ML models and the instructor has taught them both in Python as well as R. So this course will be useful in the future too. The best way to learn ML models which have worked for me is by watching and practicing the syntax using the above-mentioned course and check out the intuition part from Krish Naik’s YouTube channel.

These resources are pretty much enough to get you running for Machine Learning. Data science includes Machine Learning, then Deep Learning, Neural Networks, and AI. So naturally, Machine Learning is the starting point for your journey. I would also suggest checking out some specialization courses from Coursera but beware those specializations are quite expensive. I have spent a lot of time exploring ML and the above-mentioned resources are very helpful. I’ll be posting about other topics as and when I start learning them. For the time being, you may call this post a beginner’s guide for Data Science.

Well, remember I said that you will need to learn “a ton of things”? This is just the programming part of Data Science. You will have to eventually learn about Model Deployment, Microsoft Excel, visualization tools like Power BI or Tableau, Web Scraping etc. But, everything has its own time and there is no need to panic and rush through everything. You will have to dedicate a lot of time to this if you are planning to step into Data Science. Each topic is quite vast and different from the other. The most important thing is practice. Each time you learn a new concept, or a new ML model, browse through Kaggle and pick a dataset and apply your newfound knowledge. It will give you a much-needed boost. Even if you fail or are stuck, watch some videos on YouTube and read other’s code on GitHub to understand what mistake you are doing.

That’s all for this introduction to Data Science. I hope you guys enjoyed reading and share this post with your friends who are planning or thinking about Data Science!