Press "Enter" to skip to content

Category: R

Working with Network Graphs in R

Published 2020-10-28 by Kevin Feasel

John MacKintosh shows us the visNetwork package:

I’ve long been hoping for a reason to have to devote time to learning how to produce network plots. In my world, where bar and line charts reign supreme (with heatmaps and waffle charts thrown in occasionally) it is nice to be able to develop a new visualisation.

I’ve been wanting to produce a network plot for some time. But, the data structure, with its nodes and edges, and seeming lack of any identifiable characteristics, has meant it has not been hugely far up my agenda, or at least, never far up enough to make me learn more about it.

Click through for an example of where a network diagram can work out. H/T R-Bloggers

Comments closed

Shortest Path Calculations with Dijkstra’s Algorithm

Published 2020-10-14 by Kevin Feasel

Holger von Jouanne-Diedrich takes us through Dijkstra’s algorithm for shortest path calculations:

This post is partly based on this essay Python Patterns – Implementing Graphs, the example is from the German book “Das Geheimnis des kürzesten Weges” (“The secret of the shortest path”) by my colleague Professor Gritzmann and Dr. Brandenberg. For finding the most elegant way to convert data frames into igraph-objects I got help (once again!) from the wonderful R community over at StackOverflow.

Dijkstra’s algorithm is a recursive algorithm. If you are not familiar with recursion you might want to read my post To understand Recursion you have to understand Recursion… first.

Click through for an implementation in R.

Comments closed

Sliding Windows in R

Published 2020-10-13 by Kevin Feasel

Bryan Shalloway shows off some new functionality in the rsample package:

For some problems you may want to take a traditional regression or classification based approach while still accounting for the date/time-sensitive components of your data. In this post I will use the tidymodels suite of packages to:

– build lag based and non-lag based features
– set-up appropriate time series cross-validation windows
– evaluate performance of linear regression and random forest models on a regression problem

For my example I will use data from Wake County food inspections. I will try to predict the SCORE for upcoming restaurant food inspections.

Click through to see it in action.

Comments closed

Gussying Up R Tables in GitHub

Published 2020-10-13 by Kevin Feasel

Laura Ellis solves a problem:

One thing I love about performing analysis in .Rmd (R Markdown) files is how easy it is to share your results publicly on GitHub. Create your analysis in the .Rmd file, set your output variant as below, knit to .md format and then add your files to GitHub!

There is only one problem with the .md output: PRETTY TABLES! Most of the pretty tables packages that I like to use, or don’t display all of the formatting, or don’t display at all in .md format.

Click through to see how to solve this, including demonstration videos.

Comments closed

Time Series Forecasting in R

Published 2020-10-02 by Kevin Feasel

Selcuk Disci contrasts a couple of methods for time series forecasting:

It is always hard to find a proper model to forecast time series data. One of the reasons is that models that use time-series data often expose to serial correlation. In this article, we will compare k nearest neighbor (KNN) regression which is a supervised machine learning method, with a more classical and stochastic process, autoregressive integrated moving average (ARIMA).

We will use the monthly prices of refined gold futures(XAUTRY) for one gram in Turkish Lira traded on BIST(Istanbul Stock Exchange) for forecasting. We created the data frame starting from 2013. You can download the relevant excel file from here.

Click through for the demonstration. H/T R-Bloggers.

Comments closed

SQL Server R and Python Language Extensions Now Open Source

Published 2020-09-25 by Kevin Feasel

The SQL Server team has an announcement:

Previously, we announced a Java extension. Today, we are sharing that we are open sourcing the R and Python language extensions for SQL Server for both Windows and Linux on GitHub.

These extensions are the latest examples using an evolved programming language extensibility architecture which allows integration with a new type of language extension. This new architecture gives customers the freedom to bring their own runtime and execute programs using that runtime in SQL Server, while leveraging the existing security and governance that the SQL Server programming language extensibility architecture provides.

Very interesting.

Comments closed

Random Numbers in R: Parallel Processing Edition

Published 2020-09-23 by Kevin Feasel

Henrik Bengtsson takes us through an interesting issue:

R does a superb job of taking care of us when it comes to random number generation – as long as we run our analysis sequentially in a single R process. Formally R uses the Mersenne Twister RNG algorithm [1] by default, which can we can set explicitly using RNGkind("Mersenne-Twister"). However, like many other RNG algorithms, the authors designed this one for generating random number sequentially but not in parallel. If we use it in parallel code, there is a risk that there will a correlation between the random numbers generated in parallel, and, when taken together, they may no longer be “random enough” for our needs.

The post covers how the future package has your back when it comes to random numbers. H/T R-Bloggers.

Comments closed