According to data from Glassdoor, data scientists make an average of $117,345 a year, with starting salaries ranging from $50,000 to $90,000. With everyone from Facebook and Microsoft to Uber and Airbnb constantly looking for talented data scientists, undoubtedly here’s where the money’s at. So if you have a knack for numbers and looking to capitalize on the massive growth in this industry, you too could kick start a career in data science today. The first thing you need is an aptitude for math. In fact, most data scientists possess advanced degrees in mathematics, statistics or other related fields.

The next thing you need is a level of proficiency in computer programming, primarily working around languages that help compute all that data and make sense of it. Now this is where Python comes in. R and Python are the most commonly used programming languages for data science, out of which Python is the one that’s largely preferred by professionals. The thing about Python is that it is a general purpose back-end programming language.  It is easy to learn and get started with. It has a marked proficiency in analytical and quantitative computing. You can tell that this is one language you can count on when you hear that it has been used by the likes of YouTube and Google.

Python in Data Science

Some of the reasons Python is highly sought after when it comes to data science are:

So now you know that learning Python can be the ultimate boost you’ll need to get your data science career established for success. So today, let’s talk about how you can learn Python and leverage it to build a rewarding career for yourself. Here’s all the information you’ll ever need, to get a grip on what you need to learn, where to learn it from and how to put it to good use.

Step 1: Learn Your Way around Python

So you already know the perks of learning python, so it’s time to get cracking. There are plenty of paid as well as free courses you can take online, as well as several print books you could buy. Often, sifting through the sources to pick one is the bigger challenge than finding a way to learn Python for data science.
You can begin with a free online course for beginners in Python, such as Introduction to Python for Data Science.

Another option is the Google Class for Python that is a 2 day series you can take at any time. It makes it easy to set up Python on your computer and start with easy first steps. For instance, it shows you how to easily check if Python is installed on your system by entering

~/google-python-exercises$ python hello.py
Hello World
~/google-python-exercises$ python hello.py Alice
Hello Alice

If you find that python is not installed, go to the Python.org download page. To run the Python interpreter interactively, simply type python in the terminal:

~/google-python-exercises$ python
Python 2.5.2 (r252:60911, Feb 22 2008, 07:57:53)
[GCC 4.0.1 (Apple Computer, Inc. build 5363)] on darwin
Type "help", "copyright", "credits" or "license" for more information.
>>> 1 + 1
2
>>> you can type expressions here .. use ctrl-d to exit

For Google’s Python Class, it’s best to use Python 2.7. Although Python 3.x is becoming more popular, this course is designed for Python 2.6 or later. Once you have that, Google makes it fairly easy to learn and get started with your first Python project.

Introduction to Python is another interactive course that lets you learn at your own pace.

For those of you who like to jump into it and learn by doing, you should check out the Python Jumpstart by Building 10 Apps which is a highly interactive video course taught by Michael Kennedy.

Now, the important thing here is that you shouldn’t get too caught up in learning Python and focus on becoming a true blue Python expert. It is more important to get generally comfortable with the language and do enough practice to learn your way around it. Once you’ve done that, it is best to move on to the next step instead of lingering too long on the aspects of language. Instead, apply what you learn and focus on building data structures, data types, loops, comparisons, imports, functions, conditional statements, comprehensions and a dozen other things that pertain to data science in particular.

Step 2: Get Comfortable with Python’s Data Science Libraries

Having learnt the ropes of Python, it’s time to learn the strings of Python’s data science libraries. These help make complex tasks easy and take most of the mundane coding off your hands. The result is better models with faster, lesser code and more time spent on innovating your model and solving problems. Learn your way around the biggest, most prominent ones first, such as NumPy and SciPy for scientific and numeric computation. These two provide you with several pre-compiled functions and data structures for quick and efficient computation.

Then there is PANDAS, the Python Data Analysis Library that helps you add data structures for practical analysis and works smoothly even with incomplete, unorganized, messy and unstructured data. Another wildly popular library called Matplotlib helps with the visualization aspect by producing quality figures like plots, histograms, scatterplots, bar charts and more and even lends hardcopy formats to present interactive data. There are numerous other libraries that will help augment the power of Python in developing robust data science applications. You can check out this post by KDnuggets for a detailed anthology of all fantastic data science libraries for Python.

Step 3: Get To Work And Practice. Practice, Practice

Too much learning is no good without an equal and simultaneous amount of doing. So pick a project you’re comfortable with and put what you learnt to good use. Drive and motivation along with technical soundness are of essence here. To keep yourself motivated as well as updated, stay current on the top community blogs and forums. Attend meetups and socialize with the Python data science community to keep updated.

Entering Kaggle competitions is a great way to not only challenge yourself and learn awesome new stuff but also, any victories will add to your portfolio in a big way. The great thing about Kaggle competitions is that you don’t have to hit the familiar roadblock of not having an amazing idea. These competitions keep the idea stream running so you can just keep coding.

Start your own data science project. It doesn’t have to be hugely ambitious at this point. Choose a fitting challenge for the stage of learning you are at.

Follow blogs like DataTau and email newsletters like Data Elixir, Python Weekly and Data Science Weekly.

Step 4: Build a Portfolio

Everything you develop, be it for a Kaggle competition or a personal project can go into a portfolio for potential employers to see. As long as you can show finesse with several datasets, deliver insights and show your desire to learn as well as your skill, your portfolio doesn’t need a particular theme. Show your diversity and proficiency. Everything from a small practice project to a prize winning entry can go into your portfolio. This could easily double as your resume when applying for data science jobs down the line. Show your familiarity with new data structures, bigger statistics, insights and models.

Step 5: Advance Your Skills

One lifetime is probably too small a thing to learn all of data science. That is why it is important to make every project worthwhile. To truly build a rewarding career that keeps going higher, learn advanced data science techniques. Learn clustering, regression, classification and stay on top of emerging trends in data science. Always keep learning and experimenting.

Conclusion

Data science is undoubtedly one of the most exciting careers of our generation. With the amount of technology that surrounds us from smart devices to IoT appliances, data is truly overflowing. Companies pay top dollars to unravel trends about customer behavior from this data. If you aspire to make big on this trend, the above steps will help set a solid foundation on which you can build a rewarding career in data science. Just keep on learning and never stop applying what you learn.

Data ManagementTips & Tricks