R and Python, the “dynamic duo” of data science, are both free, open-source programming languages. That means that there’s no “vendor” in the sense that, say, Microsoft owns Excel. This can make getting started with these programs a little trickier: there are several ways to install them, often multi-step, confusing, and resource-intensive.
It would be easy as a brand-new programmer to give up on tools that are so involved even to install — “If that’s hard, just imaging trying to use them!”
Fortunately, free cloud-based applications exist for you to experiment with these programs, no installation needed. This saves you disk space and headaches and allows you to dig into the code — and the possibilities — rather than the logistics.
For R: RStudio Cloud
RStudio Cloud comes from RStudio, vendor of the predominant RStudio integrated development environment. (I use RStudio in teaching my R course.)
Simply create an RStudio account and get started. You can create a new project and run a session of RStudio from your browser. The code will execute on RStudio servers.
Your initial workspace will look like the below. This is a “virtual” instance of the RStudio interface:
If this is the first time you have worked in RStudio, check out my “Tour of RStudio” below.
To continue dabbling with R, check out my posts. Your R session will run just as it would on your computer, but this time RStudio takes care of the software.
Ready to take the plunge into R? Get started with my course, R Explained for Excel Users.
For Python: Google Colaboratory
Google hosts the free Colaboratory service for running Python using a modified Jupyter notebook. The exact “look and feel” of Colab will not be the same as using a code editor like PyCharm (my favorite environment for working in Python) or even a “plain” Jupyter notebook, but the functionality is there, plus you don’t have to deal with maintaining the software and packages.
To access Colab, log into your Google account and check out the Google Colab starter notebook, which includes the below video.
Google Colab gives you direct access to Google’s supercomputers — you can do some pretty serious data on here, as the endorsement from TensorFlow suggests (that is a popular package for deep learning built by developers at Google). You can even execute on your Google Drive files entirely from the cloud.
Conclusion: Get coding fast
If you’d like more practice getting into R and Python via RStudio and Jupyter Notebooks, with the experiences of an Excel user particularly in mind, check out my book Advancing into Analytics: From Excel to Python and R.
Thanks for the article George. RStudio Cloud is great for learning, but the 1GB of memory is very constraining. Of course you can pay for the pro version or deploy your own server. I wanted to add that you can in fact run R on colab, instructions here:
https://stackoverflow.com/questions/54595285/how-to-use-r-with-google-colaboratory
Thanks for pointing that out Rupe as I had discovered that after writing this ¯\_(ツ)_/¯. Great resource!
I didn’t know it was hard to install R on your own computer! So far I have always downloaded R from CRAN and simply installed it. And then immediately used! With Python it may be different, here I have already experienced strange things. The Anaconda distribution is already helpful there….
Thanks for reading Gunter! I think the tricky bit for at least R is indeed that two-step process, if you are not familiar with how open-source works it can be odd to need two downloads plus whatever packages. I like with RStudio Cloud you can get straight into working with the code with new learners rather than just troubleshooting installations.
The Leibniz Institute for the Social Sciences also have a project called GESIS Notebooks, which lets you practice R, Python and Julia online. Azure Notebooks also support F#.
Beautiful, thanks so much for the resource Christian!