Category: R

Dave Mason is dipping his toes into the R waters:

I think my first exposure to R was at PASS Summit 2016. Since then, I’ve made an effort to attend R sessions at SQL Saturdays. The one commonality I seem to find in all of them is a demo with (or mention of) the dplyr package. It’s a package of functions that manipulate data in data frame objects (think of them as SQL Server/relational tables…or if you’re a .NET developer, a System.Data.DataTable object). R feels inexorably tied to dplyr at this early stage for me. R is probably way more vast than I realize, but what would it be without dplyr? Would it still be as popular? Would it still be as powerful?

What’s It Good For

I’m not sure if I’m perceiving this the right way yet, but dplyr sure feels a lot like LINQ, a .NET Framework technology that provides query-like capability for C#. For instance, you can select a subset of objects from an array, sort them, find a minimum or maximum, etc. It’s kind of like querying SQL Server, just without SQL Server.

I like the comparison of dplyr against LINQ, as they’re both data querying and transformation tools whose motif is a series of functions chained together.

David Smith tells an interesting story about a reproducibility problem with data analysis:

Timo Grossenbacher, data journalist with Swiss Radio and TV in Zurich, had a bit of a surprise when he attempted to recreate the results of one of the R Markdown scripts published by SRF Data to accompany their data journalism story about vested interests of Swiss members of parliament. Upon re-running the analysis in R last week, Timo was surprised when the results differed from those published in August 2015. There was no change to the R scripts or data in the intervening two-year period, so what caused the results to be different?

The version of R Timo was using had been updated, but that wasn’t the root cause of the problem. What had also changed was the version of the dplyr package used by the script: version 0.5.0 now, versus version 0.4.2 then. For some unknown reason, a change in the dplyr package in the intervening package caused some data rows (shown in red above) to be deleted during the data preparation process, and so the results changed.

Click through for the solution, which is pretty easy in R.

Chaitanya Sagar shows a few methods for parallelizing code in R:

Parallel programming may seem a complex process at first but the amount of time saved after executing tasks in parallel makes it worth the try. Functions such as lapply() and sapply() are great alternatives to time consuming looping functions while parallel, foreach and doParallel packages are great starting points to running tasks in parallel. These parallel processes are based on functions and are also modular. However, with great power comes a risk of code crashes. Hence it is necessary to be careful and be aware of ways to control memory usage and error handling. It is not necessary to parallelize every piece of code that you write. You can always write sequential code and decide to parallelize the parts which take significant amounts of time. This will help in further reducing out of memory instances and writing robust and fast codes. The use of parallel programing method is growing and many packages now have parallel implementations available. With this article. one can dive deep into the world of parallel programming and make full use of the vast memory and processing power to generate output quickly. The full code for this article is as follows.

If you’re using Microsoft R server, there are additional parallelism options. H/T R-Bloggers

John Mount wants us to think about part-time users:

The second point I think is particularly interesting. It means:

An R user who does not consider themselves an expert programmer could be maintaining code that they understand, but could not be expected to create from scratch.

Or:

Let’s have some sympathy for the part-time R user.

This is the point we will emphasize in our new example.

Read on for a particular example.  I think this is good advice to generalize:  write your code to make it as easy as possible for “part-time” users.  This applies to custom code you write as well, as unless you are constantly in a particular part of the code base, you’ll forget the details later and have the same problems that a part-timer would have working with a different language.

David Smith looks at the odbc package in R:

The odbc package is a from-the-ground-up implementation of an ODBC interface for R that provides native support for additional data types (including dates, timestamps, raw binary, and 64-bit integers) and parameterized queries. The odbc package provides connections with any ODBC-compliant database, and has been comprehensively tested on SQL Server, PostgreSQL and MySQL. Benchmarks show that it’s also somewhat faster than RODBC: 3.2 times faster for reads, and 1.9 times faster for writes.

Sounds like odbc lets you run ad hoc queries and also lets you use dplyr as an ORM, similar to using Linq in C#.