First tour of R & RStudio

# 00 - First tour of R & RStudio

## Data Science with R &#183; Summer 2021

### Uli Niemann &#183; Knowledge Management & Discovery Lab

#### [https://brain.cs.uni-magdeburg.de/kmd/DataSciR/](https://brain.cs.uni-magdeburg.de/kmd/DataSciR/)

---

## What is `R`?

.pull-left70[
- **Statistical programming language** with focus on **reproducible data analysis**
- **Free, open source** and **available for every major OS**
- Since January 2017 more than **10,000 packages** available
  - Comprehensive **statistics** and **machine learning** packages
  - Elaborate packages to create aesthetically appealing **graphics** and **charts**

## What is RStudio?

- Open source **Integrated Development Environment (IDE)** for `R`
- First release in 2011
- Two major versions: **RStudio Desktop** and **RStudio Server**
- Download & Installation of `R` and RStudio: <https://www.rstudio.com/products/rstudio/download/>

]

]

---

## RStudio IDE

---

## R packages &#x1F4E6;

- Base `R` contains functions that are needed by the majority of users.
- Additional functions can be used on demand by loading **packages**.

A **package** is a **collection of functions, datasets and documentation** that extends
the capabilities of base R.

]

---

## Packages on CRAN<sup>1</sup>

<sup>1</sup> [CRAN](https://cran.r-project.org/), the _The Comprehensive R Archive Network_, consists of multiple worldwide mirror servers, used to distribute `R` and `R` packages.  
Figure source: <https://gist.github.com/daroczig/3cf06d6db4be2bbe3368>
]

---

## Tour: R and RStudio

---

## Calling functions

Basic function call scheme:

```r
some_function(arg_1 = val_1,
              arg_2 = val_2,
              ...)
```

]

Example: the `mean()` function:

- `x` is the only mandatory argument
- arguments `trim` and `na.rm` have default values

```r
x <- 1:10
mean(x) # trim = 0 and na.rm = FALSE
```
{{content}}

]

```
## [1] 5.5
```
{{content}}

```r
x <- c(1:10, NA)
mean(x)
```
{{content}}

```
## [1] NA
```
{{content}}

```r
mean(x, na.rm = TRUE) # NA's will be ignored
```

```
## [1] 5.5
```
{{content}}

```r
mean(x, TRUE) # match unnamed args to their position
```
{{content}}

```
## Error in mean.default(x, TRUE): 'trim' must be numeric of length one
```
{{content}}

---

## Downloading, installing and loading packages

```r
# A package has to be installed only once:
install.packages("dplyr") # -> install package "dplyr" from CRAN
# Load the package once per session:
library(dplyr)
```

Alternatively, you can install and load packages in RStudio using the Packages
tab.

To install a package that is hosted on GitHub use `remotes::install_github("<REPOSITORY>")`.

]

---

## Data Frames

- `data.frame` is `R`'s data structure for a **table**.
- A data frame is a rectangular collection of data, arranged by **variables** (**columns**) and **observations** (**rows**).

Columns of data frames are accessed with `$`:

```r
dataframe$var_name
```

---

## The Tidyverse

]

Quote from the [Tidyverse website](https://www.tidyverse.org/):

"**R packages for data science.** The tidyverse is an **opinionated collection of R packages designed for data science**. All packages share an underlying **design philosophy, grammar, and data structures**."

]

&rarr; collection of open-source `R` packages mainly for data wrangling and visualization  
&rarr; shared conventions and common APIs across all Tidyverse packages

]

---

## R Markdown

**R Markdown** is a file format to combine **code**, the associated **results** 
and **narrative text** in a simple text file, to create **reproducible reports** 
which can be **flexibly distributed in multiple ways**.

An R Markdown document is saved as `.Rmd` file. It contains both the (`R`) code 
and a prose description of a data analysis task. The R Markdown document can be 
rendered as HTML, PDF, Word and various other output formats.

**Pros** of R Markdown documents:

- **reproducibility** (reduce **copy&paste**)
- **simple Markdown syntax** for text
- a **single source document** that can be rendered for different 
target audiences and purposes
- **simple, future-proof file format** that can be managed by a version control 
system like Git or SVN

]

<https://rmarkdown.rstudio.com/>

]

???

- file extension
- Jupyter notebooks in Python (combine text, code and results)
- flexibility in terms of type of output document
- flexibility in terms of target audience -> parametrized reports
  - you: lab notebook containing code, conclusions and reasoning behind your analysis
  - decision makers: hide code, focus on charts and conclusions
  - colleagues: focus both on conclusions and code to ensure reproducibility and to facilitate collaborating
- rmd is plain text file so it can be managed by a version control system
- simple syntax

---

???

- process: Rmd -> md -> rendered document
- rmd: md + r-code
- knitr: run code and combine these code results (charts, models, console output) with md markup of rmd file
- pandoc (general file converter) converts this augmented md file to the desired output HTML, pdf, word etc.
- simplified process - based on the output, more packages are needed
- e.g. blogdown: Hugo required (open-source static site generator)

---

???

- YAML header: meta data
- code chunks: can be R but there are other languages supported, such as python, javascript, sql, ruby
- Markdown text
- inline code: allows for dynamic content (e.g.how many records) using simple commands
  - no need for manual copy and paste of the code output into the narrative text.

---

## Tour: R Markdown

---

## Structure of an `.Rmd` file

]

- **YAML header** (metadata)
  - enclosed by lines with three dashes
  - collection of key-value pairs separated by colons
- narrative text as **Markdown markup**
- **`R` code**
  - as code chunks surrounded by <code>&#96;</code><code>&#96;</code><code>&#96;{r}</code> and <code>&#96;</code><code>&#96;</code><code>&#96;</code>
  - as inline code surrounded by <code>&#96;r </code> and <code>&#96;</code>
]

---

## R Markdown help

**Markdown Quick Reference**:  
Help &rarr; Markdown Quick Reference

]

**R Markdown Cheat Sheet**:  
Help &rarr; Cheatsheets &rarr; R Markdown Cheat Sheet

]

---

## Environments

&#x26A0;&#xFE0F; Each R Markdown document has its own environment.

&rarr; Consequence: Objects from the global environment are unavailable in the .Rmd document when knitting.

]

#### Example:

1\. Run the following code in the console:

```r
x <- 5
x * 3
```

&#x1F60A; _All good!_

2\. Add the following code within a code chunk to your .Rmd file and knit the document:

```r
x * 3
```

&#x1F62D; _Knitting fails!_

---

## How will we use R Markdown?

- Every exercise comes with an R Markdown document which contains some code scaffolding.
- You submit your project report as an R Markdown document.
- The course slides are made with R Markdown.

???

- amount of scaffolding will decrease over the course of the semester

---

# Thank you! Questions?

&nbsp;