Brief introduction in R
First we have to set the R scene: What is R, how did it evolve, and what is the difference between
the R distribution found on CRAN versus the free version of R shipped by Microsoft: Microsoft R Open (MRO).
Then we take a brief look at different tools for developing R code, and we take a more closer look at installing RTVS: R Tools for Visual Studio.
- Brief history
- R distributions: CRAN versus Microsoft R Open (MRO)
- Installation of R
- Tools and resources
- Installing R Tools for Visual Studio
Using R in Visual Studio
Once everything is installed, we can get started running R scripts in Visual Studio. First we have to get used to the editor: how do we work with R projects, how can we manage the R packages we want to use, debug code etc. Then we take a look at MarkDown, which allows us to create HTML, Word or PDF reports which mix text with the result of R code, an easy way to embed R in reports.
Finally we start diving a bit deeper in the R language itself, looking at data types and apply them to practical problems, e.g. how can we easily generate random data in R, or remove null values (called 'NA' in R) from the input. We also take a look at how to load data from cvs files into R.
- Getting started with R Tools for Visual Studio
- Creating an R project and R scripts
- Intellisense and code snippets
- Package manager
- Using the variable explorer
- Debugging R scripts
- Creating reports with MarkDown
- R Data types: numeric, integer, logical and character
- R Data structures: vectors, factors, lists, matrices and data frames
- NA: working with nulls
- Indexing vectors and data frames
- Loading data with read.csv
R in Power BI
Power BI has become a very popular BI tool, both for business users (with the free Power BI Desktop app)
but it can also be used to distribute enterprise reports using the PowerBI.com service. For those not familiar with Power BI we first show you around in Power BI. Then we focus on the three different ways we can use R in Power BI: as a data source, a transformation or for plotting the data. In each of these domains we do not only focus on how to apply this functionality in the Power BI Desktop tool, but we also discuss the functions and packages needed in R to do data loading, transformation and plotting in general.
- Introduction Power BI
- R as a Power BI data source
- R as a Power BI transformation
- Data manipulations in R: Filtering, sorting, aggregating and joining data frames
- Using the dplyr and reshape R package
- R as a Power BI visualization
- Plotting with plot, abline, hist and other functions
- Introducing the ggplot package
- R in PowerBI.com
R in SQL Server 2016 and later
Copying data stored in SQL Server tables into an R variable is easy to do. But what if the data volume
becomes too large, or you want to run your script multi-threaded? This is where Microsoft R Server
comes into play. In this module you learn how it is different from the 'regular' R we saw so far and
the two different ways to work with R Server in SQL Server 2016 or 2017 (as a data scientist and as an application developer). But we can get the advantages of Microsoft R Server also when the data is not stored in SQL Server: this is where xdf files come into play.
In this module we also introduce how to use machine learning in R.
- Microsoft R Server (MRS) versus Microsoft R Client (MRC)
- Using the RevoScaleR package
- Running external R scripts in SQL Server
- Using R code via sp_execute_external_script in Reporting Services
- Working with Xdf files
- Machine learning in R
R in Azure
Not only on premise but also in the Microsoft cloud R can be used in lots of different ways. In this module we first introduce Azure Machine Learning (AzureML) and see how we can use R Scripts and R models in this application. AzureML also provides Jupyter (JUlia, PYThon and R) notebooks in which we can interactively run R code on our data. Azure also provides Jupyter notebooks independant of the AzureML service.
Finally we take a brief look at the Hadoop stack in Azure, HDInsight, on which we can use the Microsoft R Server technology as well.
- Introducing Azure Machine Learning
- R Scripts in Azure Machine Learning
- Create your own R models in Azure ML
- Jupyter notebooks
- Notebooks in Azure
- R Server in HDInsight
The statistical language R has become very popular as a tool for the data scientist. Microsoft has made major steps in
integrating R in its business intelligence stack. The learning goals of this course are twofold:
- Introduce students into the langauge R in general
- Show students how they can use this R language in the most important Microsoft BI tools: SQL Server, Reporting
Services, Power BI, Azure and Visual Studio
This course assumes no background in R, but a general background in the Microsoft business intelligence stack can be useful.