Fundamental of Data Science for EESS

# Fundamental of Data Science for EESS

## R session 01 - Introduction to R

2019-01-17

.pull-left[
<img src="img/NTU-Logo-full-colour.png" width="50%" style="display: block; margin: auto auto auto 0;" />
]

---

# Outline

.font150[
* What is R and why use R ?
* Resources
* Get started
* Fundamentals of R
    - Data objects
    - Vectors
    - Operators
    - Functions
    - Packages
    - Data frames
]
---
layout: true

# Introduction

---

- .font150[Who has used R before ?]

- .font150[What other programming language have you used before ?]

- .font150[For those who are experts in R]

--
    * please refrain to answer during this session...
    * help your neighbor...

---

## Computer languages

---

## History of R

* **Mid 1970s** - S Language for Statistical Computing conceived by John Chambers, Rick Becker, Trevor Hastie, Allan Wilks and others at Bell Labs

* **Early 1990's** - R was first implemented in the early 1990’s by Robert Gentleman and Ross Ihaka, both faculty members at the University of Auckland.

* **1995** - Open Source Project

* **1997** - Managed by the R Core Group

* **2000** -  First release of R

* **2011** - First release of R studio

* [Historical notes - Paper from 1998](https://www.stat.auckland.ac.nz/~ihaka/downloads/Interface98.pdf)

---
## Why use R ?

- **Script vs. Menu driven software (e.g. Excel)**
     + Can be re-rerun with new data
     + Reproducible workflow
--
- **Open source**
    + Huge number of libraries
    + Tidy "universe" : tidyverse and ggplot2
        + Very easy to manipulate tables (select columns, create new variables)
        - High quality graphics
--
- **Work environment**
    - R studio
--
-  **Document your data processing**
    - R markdown
    - Create HTML, pdf, presentations
--
- **Share your data and workflow**
    - GitHub

---
## What can you do with R ?

- **Science**
    * Statistics of course...
    * Data processing
    * Graphics
    * Time series analyses
    * Maps
    * Bioinformatics
]

- **But also**
    * Teach
    * Do a presentation
    * Write your CV
    * Build a web site
    * Write a book
    * Much more...
]

--
.center[
<img src="img/web-site-dv.png" width="50%" style="display: block; margin: auto;" />
]

---
layout: true
# Resources

---

## Books and Manuals
.center[
<img src="img/R_nutshell.png" width="30%" /><img src="img/R_graphics_cookbook.png" width="35%" />
]
* [R intro](https://cran.r-project.org/doc/manuals/r-release/R-intro.pdf) : Very good introduction to R, short and clear
* [R in a nutshell](http://rbasicsworkshop.weebly.com/uploads/1/8/6/0/18603232/adler_2009_r-inanutshell.pdf) : Many many receipes to solve all your questions
* R graphics cook book : very good for graphics

---

## On line courses and web sites

* [Coursera](https://www.coursera.org/learn/r-programming)
*	[Pluralsight](https://www.pluralsight.com/courses/r-programming-fundamentals)
* [Quick-R, very simple](http://www.statmethods.net/)

---

## Cheat sheets

* [R basics](http://github.com/rstudio/cheatsheets/raw/master/base-r.pdf)
* [ggplot2](https://github.com/rstudio/cheatsheets/raw/master/data-visualization-2.1.pdf)
* [dplyr](https://www.rstudio.com/wp-content/uploads/2015/02/data-wrangling-cheatsheet.pdf)

---

## Forum

* https://stackoverflow.com/
*	http://r-statistics.co/
*	https://stats.stackexchange.com/
*	https://www.r-bloggers.com/

---

---

## Setup

* Install [R](https://cran.r-project.org/index.html)

* Install [R studio](https://www.rstudio.com/products/rstudio/download/#download)

---

## The R studio interface

- **Bottom left**
    - Console
- **Top left**
    - File editor 
- **Top right**
    - Environment (i.e. R objects)
    - History
- **Bottom right**
    - Files
    - Plots
    - Packages
    - Help
]

.pull-rigth[
<img src="img/R_Studio_interface.png" width="50%" style="display: block; margin: auto;" />
]

---

## Create a new project

* Open R studio
* Create new project for the course in a new directory
    - e.g. Data Science Class

---

## Your first script

```r
print("Hello world")
```

```
[1] "Hello world"
```

### Two ways to proceed

1. Type directly in command window

2. Create a new script

---

---
## Everything in R is an **object**

* Assignement done with **<-**

```r
> x <- 1
> y <- 2
> x + y
```

```
[1] 3
```
--

```r
> z <- x + y
> z
```

```
[1] 3
```

---

**=** can be used instead of **<-** but refrain from it (not good style)

```r
> z = x + y
```

You can view the values of the objects in R-studio environment window (top-right)

---

## R is **case sensitive**

```r
> Z
```

```
Error in eval(expr, envir, enclos): object 'Z' not found
```

---

## Rules for naming objects
* Use
    * letters
    * numbers
    * the dot
    * the underscore (not the minus sign !)
    
    
* Start always with a letter
    * `Myvariable`, `Myvariable1`, `Myvariable.1`,`Myvariable-01` are OK
    * `1Myvariable`, `My-variable`, `Myvariable@` are **not** OK

---

## Use consistent naming
Five conventions

* alllowercase: e.g. adjustcolor
* period.separated: e.g. plot.new
* **underscore_separated**: e.g. numeric_version
* lowerCamelCase: e.g. addTaskCallback
* UpperCamelCase: e.g. SignatureMethod

Prefer third one, much more easy to read
* Use **names** for objects : **last_name**
* Use **verbs** for function : **build_name**
* Think about best order
    - e.g. prefer maybe **name_last** because then you can have name_first, name_full... 
    - and you identify that all these objects are related to a name...

---
layout: true
# R objects

---

## Data types

* **character**: "Daniel", "This is a course in R", 'Donald'

* **numeric**: 2, 15.5, 10e-3

* **integer**: 2L (the L tells R to store this as an integer)

* **date**: 2018-02-25

* **logical**: TRUE, FALSE

* **complex**: 1+4i (complex numbers with real and imaginary parts)

* **No data** "NA"

* **Not a number** "NaN" (e.g. division by zero)

---

## Data structures

* **Vector**

* **List**

* **Matrix**

* **Data frames**

* **Function**

---
layout: true
# Vectors

---

The basic R structure is a vector: 
`$$\begin{bmatrix}10 \\ 20 \\ 30 \end{bmatrix}$$`

A vector can with a single element only
`$$\begin{bmatrix}10 \end{bmatrix}$$`

--
## Assign a value to a vector

```r
x <- 10
x
```

```
[1] 10
```

---

## Assign several elements

```r
x <- c(10, 20, 30)
x
```

```
[1] 10 20 30
```
--

## Assign range

```r
x <- 10:30
x
```

```
 [1] 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
```

---

## Assign characters

```r
PoTU <- c("Donald", "Trump")
PoTU
```

```
[1] "Donald" "Trump" 
```

## Assign logical

```r
flags <- c(TRUE, FALSE, TRUE)
flags
```

```
[1]  TRUE FALSE  TRUE
```

---

## Access specific elements of a vector

### First

```r
x[1]
```

```
[1] 10
```

### Range

```r
x[1:5]
```

```
[1] 10 11 12 13 14
```

### Remove one element

```r
x[-1]
```

```
 [1] 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
```

---

## Determine object properties

Apply functions (we will come back to functions latter)

* **typeof()** - what is the object’s data type (low-level)?
* **length()** - how long is it? What about two dimensional objects?

```r
typeof(x)
length(x)
```
--

```
[1] "integer"
```

```
[1] 21
```

---
layout: true
# Operators

---

## Arithmetic Operators

| Operator |	Description | 
|---|---|---|
| +	| addition|   
| -	| subtraction |  
| *	| multiplication| 
| /	| division | 
| ^ or ** | 	exponentiation | 
| x %% y | 	modulus (x mod y) 5%%2 is 1 | 
| x %/% y	| integer division 5%/%2 is 2 |

---

## Arithmetic Operators

We are performing vector operations !

`$$\begin{bmatrix} 1\\2\\3\\..\end{bmatrix}+\begin{bmatrix}1\\2\\3\\..\end{bmatrix}=\begin{bmatrix}2\\4\\6\\..\end{bmatrix}$$`

---
## Arithmetic Operators

Vector one element

```r
 x <- 1
 y <- 2
 z <- x + y
 z
```

```
[1] 3
```

---
## Arithmetic Operators

Vector several elements

```r
# Two instructions on the same line
 x <- 1:9;  y <- 1:9
 z <- x + y
 z
```

```
[1]  2  4  6  8 10 12 14 16 18
```
--

.warning[
* Several instructions on same line separate by **;**
* The hastag **#** indicate a comment -> Use heavily to document your code
]
--
.student[
Use the other operators
]

---
## Arithmetic Operators

What happens when the vectors have different number of elements ?

```r
x <- 1:9
y <- 1
z <- x + y
z
```

```
[1]  2  3  4  5  6  7  8  9 10
```

Equivalent to

```r
y <- c(1, 1, 1, 1, 1, 1, 1, 1, 1)
```

The recycling rule...

---

## Can we add logical ?

```r
x <- TRUE
y <- FALSE
z <- x + y
z
```

```
[1] 1
```

---

## Can we add logical ?

It does not give an error but...

The resulting variable is transformed to a **numeric**

```r
typeof(x)
```

```
[1] "logical"
```

```r
typeof(z)
```

```
[1] "integer"
```

---

## Logical Operators
| Operator |	Description | 
|---|---|---|
| <	| less than	|
| <=	| less than or equal to	|
| >	| greater than	|
| >=	| greater than or equal to	|
| ==	| exactly equal to	|
| !=	| not equal to	|
| !x	| Not x	|
| x &#124; y	| x OR y	|
| x & y	| x AND y	|
| isTRUE(x)	| test if X is TRUE	|

---

## Logical Operators

```r
x <- TRUE
y <- FALSE
z1 <- x | y
z2 <- x == y
```

```
[1] TRUE
```

```
[1] FALSE
```

---

## Can we add characters ?

```r
first <- "Donald"
last <- "Trump"
full <- first + last
```

Generates an error

```
Error in first + last: non-numeric argument to binary operator
```

--
.student[
What can we do ?
]

---
layout: true
# Functions

---

Function perform specific task on objects

* e.g. to concatanate strings we use **paste0()**

```r
paste0(first, last)
```

```
[1] "DonaldTrump"
```
--

* Functions take **arguments** and return an object called **result**

* To know the arguments use ?

```r
  ? paste0()  # Do not forget the parenthesis
```
--

* Can go directly to Help panel and type function name

---

## Help

---

## Help

---
## Getting what you want

We would like to write "Donald Trump" but we have :

```r
paste0(first, last)
```

```
[1] "DonaldTrump"
```

```r
paste(first, last)
```

```
[1] "Donald Trump"
```

---
## Write your own function

```r
my_sum <- function(a, b) {
    c <- a + b
    return(c)
}
```

```r
my_sum(10, 20)
```

```
[1] 30
```

> End of lecture one
---
## Examples of functions

Most of the time you do not have to write functions because someone has already written one for what you want to do...

* Sum

```r
x <- 1:100
sum(x)
```

```
[1] 5050
```

* Normal distribution

```r
y <- rnorm(100, mean = 0, sd = 1)
y[1:10]
```

```
 [1]  0.4885882 -0.6260146 -0.8855401 -1.2341267  0.3726551  0.8956950
 [7]  0.9124247  0.1755346  0.4628793 -1.5012981
```

---

## Statistics

```r
mean(y)
```

```
[1] 0.01007783
```

```r
sd(y)
```

```
[1] 0.8875528
```
--
.student[
* Is the mean close to expected mean ?
* What can be done ?
]

---

Sample more points... 10,000 instead of 100

```r
y <- rnorm(10000, mean = 0, sd = 1)
mean(y)
```

```
[1] 0.01251073
```

```r
sd(y)
```

```
[1] 1.009886
```

---

## Plot

```r
library(graphics)
hist(y)
```
]

<img src="R-session-01-intro_files/figure-html/unnamed-chunk-58-1.png" style="display: block; margin: auto;" />
]

---

---
Packages are set of functions that have a common goal

They are really the strength of R

And these are only the "official"" packages. You can find more on GitHub

---

## Installing a package

Download on your computer the package you need

.center[
<img src="img/R_studio_package_01.png" width="45%" /><img src="img/R_studio_package_02.png" width="35%" />
]

---

## Using a package

To use functions from the package

- use the syntax `package::function`

```r
stringr::str_c(first, last, sep = " ")
```

```
[1] "Donald Trump"
```

- load the package with the library function

```r
library(stringr)
str_c(first, last, sep = " ")
```

```
[1] "Donald Trump"
```

---

## List installed packages

---

# Other objects

* List

* Matrix

* Factors

* **Data frames**

---

# Data frames

---

## What is it ?

* Table mixing different types of columns (an Excel table...)
* However within a column all values are similar, e.g. numeric, logical, character

```r
df <- data.frame(label = letters[1:6], id = 1:6, value = rnorm(6, mean = 0, 
    sd = 1), flag = c(TRUE, FALSE), stringsAsFactors = FALSE)
df
```

```
  label id      value  flag
1     a  1  0.8002749  TRUE
2     b  2 -0.1723698 FALSE
3     c  3  1.0188527  TRUE
4     d  4 -1.4748408 FALSE
5     e  5  0.5381787  TRUE
6     f  6  0.8350807 FALSE
```
.warning[
* We will not get into the factor at this time, why we set `stringsAsFactors = FALSE` 
* Notice the recycling rule ?
]

---

## Useful functions

```r
dim(df)  # returns the dimensions of data frame
```

```
[1] 6 4
```

```r
nrow(df)  # number of rows
```

```
[1] 6
```

```r
ncol(df)  # number of columns
```

```
[1] 4
```

---

## Useful functions

```r
str(df)  # structure of data frame - name, type and preview of data in each column
```

```
'data.frame':	6 obs. of  4 variables:
 $ label: chr  "a" "b" "c" "d" ...
 $ id   : int  1 2 3 4 5 6
 $ value: num  0.8 -0.172 1.019 -1.475 0.538 ...
 $ flag : logi  TRUE FALSE TRUE FALSE TRUE FALSE
```

```r
colnames(df)  # columns names
```

```
[1] "label" "id"    "value" "flag" 
```

---

## Access specific column
* Use `$`notation

```r
df$value
```

```
[1]  0.8002749 -0.1723698  1.0188527 -1.4748408  0.5381787  0.8350807
```

* Use the `df[i,j]` notation

```r
df[, 3]
```

```
[1]  0.8002749 -0.1723698  1.0188527 -1.4748408  0.5381787  0.8350807
```

```r
df[, "value"]
```

```
[1]  0.8002749 -0.1723698  1.0188527 -1.4748408  0.5381787  0.8350807
```

---

## Access specific value
* Use `$`notation

```r
df$label[5]
```

```
[1] "e"
```

```r
df$label[1:5]
```

```
[1] "a" "b" "c" "d" "e"
```

* Use the `df[i,j]` notation

```r
df[5, 1]
```

```
[1] "e"
```

```r
df[1:5, "value"]
```

```
[1]  0.8002749 -0.1723698  1.0188527 -1.4748408  0.5381787
```

---

## Filter the data

```r
df[df$id <= 3, ]
```

```
  label id      value  flag
1     a  1  0.8002749  TRUE
2     b  2 -0.1723698 FALSE
3     c  3  1.0188527  TRUE
```

```r
df[df$label == 3, ]
```

```
[1] label id    value flag 
<0 lignes> (ou 'row.names' de longueur nulle)
```

---
layout: false

# Next time: Markdown

* Mix text, R code and R output in a single document
* Produce documents as HTML, pdf or even Word from the same template
]

<img src="img/R-markdown-book.png" width="40%" style="display: block; margin: auto;" />
]

.student[
* Please install the following packages and their dependencies
    * rmarkdown (will install also knitr)
    * tinytex (Latex)

* Read the installation instruction : https://bookdown.org/yihui/rmarkdown/installation.html
* Have a look at the book https://bookdown.org/yihui/rmarkdown
]