What is R?

R is technically a language and environment that provides a lot of easy but robust functionalities in the realm of advanced/general statistical computing and it also provides a very useful platform to generate graphs or graphical representations of various analysis as well. It was actually developed by the research scientists (John Chambers and other team members) of the Bell labs. R was conceptualised in terms of a different implementation of the S (another language), though there are quite a few differences in the overall framework.

R is quite flexible in terms of its implementation of various statistical and graphical methods and this framework is also getting aligned to state-of-the-art scalable environments (like H2O etc.). In a nutshell, R provides a gateway or an open-source route to implement numerous different functionalities in it.

One great highlight of R’s capabilities would be the kind of charts or graphical representations of the data, statistical tests and many more functionalities that can be created using R. This is an evolving space and almost every day new capabilities are getting added in R. It also gives the user the full control to come-up newer approaches to experiment with as well.

R is available as Free Software under the terms of the Free Software Foundation’s GNU General Public License in source code form.

The R environment

R is an integrated suite of software facilities for data manipulation, statistical analysis, and graphical display. Some of the key highlights about what R combines:

• an effective data handling and storage capability,
• a great tool for matrix operations – Which is a requirement for any modern-day computing environment and quite useful for Machine Learning.
• It provides an integrated collection of intermediate tools for data analysis, data processing, etc.
• It provides graphical facilities for data analysis and can generate graphs for other usages.
• a very well-developed programming language which includes advanced functionalities like conditionals, loops, user-defined recursive functions and input, and output facilities, etc.

Performance

R is not typically a fast language. This is purely an outcome of choice!. R was purposely designed to make data analysis and statistics easier for you to do. It was not designed to make life easier for your computer. While R is slow compared to other programming languages, for most purposes, it’s fast enough.

Extreme Dynamism

R is an extremely dynamic programming language. Almost anything can be modified after it is implemented. To give just a few examples:

• Change the body, arguments, and environment of functions.
• Modify objects outside of the local environment with <<-.

The only thing you can’t change are objects in sealed namespaces, which are created when you load a package.

The advantage of dynamism is that you need minimal upfront planning. You can always do a course correction, iterating your way to a solution without having to start afresh. On the other hand, the cons of such dynamism are that it is difficult to predict what’s going to happen, the easier it is for an interpreter or compiler to make optimizations. (If you’d like more details, Charles Nutter expands on this idea at On Languages, VMs, Optimization, and the Way of the World.) If an interpreter can’t predict what’s going to happen, it has to consider many options before it finds the right one. 

Alternative R Implementation

There are quite a few exciting new implementations of R that are available. While they all try to stick as closely as possible to the existing language definition, they also improve speed by using ideas from modern interpreter design. The four most mature open-source projects are:

• pqR (pretty quick R) by Radford Neal. Built on top of R 2.15.0, it fixes many obvious performance issues and provides better memory management and some support for automatic multithreading.
• Renjin by BeDataDriven. Renjin uses the Java virtual machine and has an extensive test suite.
• FastR by a team from Purdue. FastR is similar to Renjin, but it makes more ambitious optimizations and is somewhat less mature.
• Riposte by Justin Talbot and Zachary DeVito. Riposte is experimental and ambitious. For the parts of R it implements, it is extremely fast. Riposte is described in more detail in Riposte: A Trace-Driven Compiler and Parallel VM for Vector Code in R.

Why R is in demand?

R was actually built by statisticians, for statisticians – and most developers can tell that by looking at its particular syntax. One major reason of popularity for R is also related to the recent developments in the realm of machine learning and deep learning. Since the mathematical computations involved in machine learning are derived from statistics, R comes in handy to those who want to gain a better understanding of the underlying details and build something truly innovative.

Let’s discuss some pros and cons of R as a go-to language for ML/DL tasks:

Pros:

• It is quite perfect for data analytics or visualization – if these are at the heart of your project, R is an excellent choice. It allows rapid prototyping and working with datasets to build machine learning models.
• It includes a great number of libraries and tools – R has plenty of packages that improve its performance in machine learning projects. 
• It is also great for exploratory work as well.

Cons:

• The steep learning curve for R
• Inconsistency – since in R algorithms come from third parties, you might end up with many inconsistencies. Every time your development team uses a new algorithm, they will also need to learn new ways to model data and make predictions.

Given its wide range of usage, it is quite important to learn the key nuggets of this language. In this tutorial, we will delve deep further and learn about various nuances of the R language.