Diagram SQL Server Graph Databases in R – Part 1

by edpflager

Welcome to the first of several articles covering generating diagrams in R of Microsoft SQL Server graph databases. A little theoretical background is in order before diving into the mechanics. In relational databases, tables typically connect or join based on a unique value in one table to another table where the same value may appear one or more times. Called a one-to-one or one-to-many relationship, it functions very well for most purposes. Dealing with real world situations with many-to-many relationship representation has been more awkward. As an example, the image above shows a representation of a classic many-to-many situation: class enrollment. A single student may be enrolled in many classes, and a single class may contain many students. In order to model this in a relational database, you would need to use an intermediary table, with one row for each student and each class they are enrolled in. By doing it this way, you can query the tables in either direction to get a list of all the students with the classes they are enrolled in, or all of the classes and which students are enrolled in those classes.

(more…)

Continue Reading

Tidyverse installation on Linux Mint

by edpflager

If you’ve been using R for even a short time, you more than likely have heard of the Tidyverse from RStudio and originally developed by Hadley Wickham.  As explained on the tidyverse website this set of packages works together with a common data representation methodology and API design.  They provide a host of functionality to clean and tidy your data to make it easier to analyze. Similar data representation between packages lets you define an entity for one operation, and use that entity through an entire workflow rather than creating intermittent entities along the way, resulting in more efficient code, and requiring fewer  system resources. The tidyverse methodology mimics good database design, with one row of data consisting of a single record or observation with variables assigned to columns in that row, and each value being assigned to its own cell. Having worked with databases for over a decade, the structure of tidy data makes sense. See Hadley Wickham’s book for more information.

Currently there are 26 separate packages in the tidyverse, and even more that inter operate with them, so they can be a big part of your R workflow. The installation of the tidyverse is fairly easy in most situations, using the standard method  in R Studio or at the R command line:

install.packages("tidyverse")

But when setting up a new Linux Mint 19.1 system, that didn’t work for me. The output included a long list of errors, and some helpful text to install this or that Debian package, buried in the text that was scrolling by. To make this process easier for Mint users, here are instructions on operating system requirements to install before attempting to install  the Tidyverse. (more…)

Continue Reading

More Control over Footnotes in R Markdown

by edpflager

In R Markdown include footnotes in documents to provide a reference or comment that may provide additional context or an aside to the topic. Traditionally, these numbered notes are placed at the bottom of the page and a corresponding superscript number is added to the text body at the relevant text. R Markdown supports footnotes natively without the need for any additional packages.I’ll cover that first, and then show how to gain more control over footnotes using an additional LaTeX package.

The native footnote syntax is simple, and requires just a paired set of footnote tags.

[^1] – denotes where the footnote superscript notation should be placed.

[^1]: {footnote text} – is the corresponding footnote text. The character within the brackets has to be match the superscript tag and be followed by the trailing the colon. This footnote text can be placed at any spot in the document, but for clarity’s sake, place it close to the text it refers to.

(more…)

Continue Reading

Quick Tips – Spacing and Bulleted Lists

by edpflager

Wrapping up the year with a few simple tips for R Markdown that work with PDF, Word or HTML output.

\VSPACE

When you insert a series of blank lines in an R Markdown document and knit the document, the parsing of your document strips out those blank lines. As an example I entered in my file the lines on the left in the image below, and when I Knitted the document, the results were those on the right:

(more…)

Continue Reading

RStudio on Linux Mint 19

by edpflager

R Studio is a popular IDE to use with R and runs on Windows (7/8/10), Mac OS X (10.6+) and the most widely used Linux distributions: Debian/Ubuntu and Fedora/Red Hat/OpenSuSE.

For Linux, that generally means installation on any other distributions that are derivatives of these should also work with with minimal effort. For my personal Linux efforts, I have bounced between CentOS (a derivative of Red Hat), and Linux Mint (version 19 built from Ubuntu 18.04). Recently I picked up a refurbished Acer Swift 1 (cheap!) so I would have a Linux laptop to work with.

The Comprehensive R Archive Network (CRAN) has Ubuntu installation packages of R-base up through Ubuntu Cosmic 18.10. Before installing RStudio, you need to install base-R the minimum installation of R. Here is instructions for that based on the CRAN website.

(more…)

Continue Reading

SmartDiagram in R Markdown

by edpflager

A useful plugin from the LaTeX environment that works with RMarkdown is smartdiagram developed by Claudio Fiandrino. Conceived as a way to create diagrams from a list of items, it will generate output as circular diagrams, flow diagrams, sequence diagrams and several others. Components within the diagrams can be different colors or a single uniform one. Combined these elements provide a visual layer to describe activities within your R Markdown diagrams and make what your are trying to convey more intuitive to the reader. I won’t cover using all of the different diagrams in smartdiagrams with R Markdown, but this should be a good introduction to how to get it to work.

To get started, in your Markdown document YAML header-includes line add a call to use the smartdiagram package:

header-includes: \usepackage{smartdiagram}

(more…)

Continue Reading