Practicing Data Science Skills on a Budget

I am often asked by my students how they can practice their data science skills without spending too much money. Not everyone has the funds for expensive equipment or software. Although there will always be a tipping point whereby you will need to “spend money to make money,” there are alternatives to buying a $1,500 laptop. And that’s is how I started.

While I love R and RStudio, I don’t always recommend them for beginners, but I do recommend free open source tools. But simply having some free tools doesn't mean you can easily download data sets and run data science projects. You need low-cost hardware as well as low-cost software.

That’s why I almost always recommend that students get a Chromebook. Currently, I use a 14” Acer Chromebook that costs $279. It has 3GB RAM—I know what you’re thinking—how can I run a data science workflow on only 3GB RAM with hardware that runs an Atom CPU?

Let’s table that question for a moment and explore some of the wonderful reasons why a Chromebook brings a lot to the table.

Why I Recommend Chromebooks

One reason I love my Chromebook is the weight, or lack thereof! They are very portable and my Chromebook is my go-to tool for hanging out in WiFi hotspot locations. It’s easy to carry, easy to secure, and easy to use.  Further, they have great battery life, and security updates are done automagically.

The Wonders of Web-based Software

There are so many online software apps these days, that I almost never need my laptop. Let’s start with Google app. I’m writing this blog in Google Docs on my Chromebook. For editing PDFs, I use Kami. And for editing images? I use Pixlr. What about file sharing, file conversion, mind maps, network diagrams, kanban, chat? I use a grammar checking tool. There’s a web-based app for just about everything. But what about data science?

Yes, Even Data Science

You can also run R and RStudio from the web, at no cost! You just need an Amazon account so that you can setup a free tier AWS instance to run a Linux server and an RStudio server. Follow these easy to follow steps to set up and you’re ready to begin. Once it’s all set up, you can connect to your RStudio environment and run R, just like that. Just remember to keep an eye on the AWS services to make sure that you don’t exceed the free tier parameters. Otherwise you’ll be charged a fee.

You can also find countless other web solutions that run in Chrome. One of my favorite sites is a regular expression editor. As you may know, regular expressions are “search” on steroids—a must tool in any data scientist’s toolbox.

The only drawback with using a Chromebook is you can’t have 40-50 open tabs running at once, because you’ll run out of memory. This is easily mitigated with a fabulous Chrome extension called The Great Suspender.

Cloud-based Practice Labs

Pressure-free practice is crucial when starting out in data science, especially at the student stage. The majority of Simplilearn’s data science training courses include labs for this very reason. Data Science Training with R, for example, includes a feature called CloudLabs that students can access directly from the learning management system throughout the duration of the course. Similarly, Python training includes Jupyter notebook labs.

So if you’re just starting out, whether with Simplilearn or on your own, you can run all of your test codes on the cloud without polluting your production environment while mastering data science.

About the Author

Peter FerrariPeter Ferrari

Peter is a data science instructor for Simplilearn whose passion is data, with over three decades of experience in information technology, mathematics, training and data science. His goal is to help educate everyone and anyone on how much more powerful data-driven decisions can be rather than the “gut-feeling”-based decisions that so many business executives have historically used and continue to use.

View More
  • Disclaimer
  • PMP, PMI, PMBOK, CAPM, PgMP, PfMP, ACP, PBA, RMP, SP, and OPM3 are registered marks of the Project Management Institute, Inc.