notes.

md 10/3/2022

R programming language
history
Robert Gentleman has developed R in 1993
Robert got inspirations from another language S [Statistics] from IBM
developed for performing the statistical calculations
features
it is not a general purpose development language
can not be used for
developing console applications
developing web applications (*)
developing mobile applications
can be used to develop applications involving the statistical calculations
provides the general purpose programming constructs
variable declarations
function declaration and invocation
loops
condition
data types
collections
provides in-built statistical function
provides various types of packages
graphical packages (ggplot)
machine learning packages
data analysis packages
data analytics packages
is a scripting language (there is no entry point function)
environment setup
install latest version of R programming language from
https://cran.r-project.org/bin/windows/base/
IDEs
RStudio [https://rstudio.com/products/rstudio/download/]
Pycharm [https://www.jetbrains.com/help/pycharm/r-plugin-support.html]
Editor
VSCode
update system environment variable named path with the bin directory of python and R
execute the code
1 / 20
notes.md 10/3/2022

using command prompt

execute the command R to get the interactive R command prompt
write and execute the R code line by line
using file (script)
write the code in a file with an extension R
use Rscript to execute the file

> Rscript test.R

shortcuts in RStudio
ctrl + l => clear the console
ctrl + shift + n => creates a new R script
ctrl + enter => run the current line or selected lines
ctrl + shift + s => runs the current entire document
? => shows help of the function name
?? => search for the topic
identifier
used to identify an entity like variable or a function
rules to declare identifier
can not start an identifier with a number
e.g. 1name is invalid
can not start an identifier with an underscore
e.g. _varname is invalid
can not use special characters like space, symbols (#$%@)
e.g.
invalid identifier (has space within it): first name
invalid identifier: name@1, name#, name%
identifier can use a special character dot (.)
e.g.
valid identifier: first.name
valid identifier: can.vote
convention
to create an identifier for first name
first_name (preferred in python)
first.name (preferred in R)
variable

a memory used to store a value

a variable can change its value
variable declaration / assignment
2 / 20
notes.md 10/3/2022

# python like syntax

num = 100

# single headed arrow syntax

num <- 100

# double headed arrow syntax

num <<- 100

# single headed arrow syntax

100 -> num

# double headed arrow syntax

100 ->> num

to print variable value on console

num = 100
print(num)

# print value of a variable along with a constant string

cat("num = ", num)
paste("num = ", num)
paste0("num = ", num)

pre-defined values
NA
Not Available
Inf
stands for infinity
data types
in R, data type is always inferred
automatically decided by R by looking at the current value inside the variable
types
Vectors
Lists
Matrices
Arrays
3 / 20
notes.md 10/3/2022

Factors
Data Frames
Vectors

represents one dimensional array

collection of similar values
data types
numeric
represents a number
e.g.

# Vector of numeric value

num = 100

# Vector of numeric value

salary = 10.15

integer
represents a whole number
e.g.

# vector of integer
age = 30L

character
represents a string
e.g.

# vector of character
first.name = "steve"

# vector of character
last.name = 'jobs'

logical
represents boolean values like
4 / 20
notes.md 10/3/2022

TRUE or FALSE
T or F
e.g.

# vector of logical
can.vote = TRUE

raw
represents raw characters
e.g.

# vector of raw
address = charToRaw('pune')

# vector of raw
hello = charToRaw('नम ार')

complex
represents a complex number
complex number contains
real part
imaginary part
e.g.

# vector of complex
complex.number = 10 + 20i

a collection of similar values

# vector of numeric
# v.1 = [10, 20, 30, 40, 50, 1]
v.1 = c(10, 20, 30, 40, 50, TRUE)

# vector of character
# v.2 = ["10", "20", "steve", "30", "40"]
v.2 = c(10, 20, "steve", 30, 40)

# vector of character

5 / 20
notes.md 10/3/2022

# v.3 = ["10", "20", "30", "40", "50", "TRUE", "india"]

v.3 = c(10, 20, 30, 40, 50, TRUE, "india")

indexing
index starts at one (1) [does not start at zero (0)]
positivie indexing
e.g.

numbers = c(10, 20, 30, 40, 50, 60, 70, 80, 90, 100)

# get the first value from the vector

# 10
print(numbers[1])

# get the last value from the vector

# 100
print(numbers[10])

# get the values at postions 2, 4 and 6

# [20, 40, 60]
print(numbers[c(2, 4, 6)])

negative indexing
exclude the index and return remaining values
e.g.

numbers = c(10, 20, 30, 40, 50, 60, 70, 80, 90, 100)

# get all the values except first position

# [20, 30, 40, 50, 60, 70, 80, 90, 100]
print(numbers[-1])

# get all the values except last position

# [10, 20, 30, 40, 50, 60, 70, 80, 90]
print(numbers[-10])

# get all the values expect from positions 2 4 and 6

# [10 30 50 70 80 90 100]
print(numbers[c(-2, -4, -6)])

slicing
6 / 20
notes.md 10/3/2022

taking a portion out of a vector

e.g.

numbers = c(10, 20, 30, 40, 50, 60, 70, 80, 90, 100)

# 60, 70, 80, 90, 100

print(numbers[-1:-5])

# 60, 70, 80, 90, 100

print(numbers[6:10])

broadcast operator
performing operation on every member of the vector
mathematical operators 1

# add 10 to every number

print(numbers.1 + 10)

# subtract 10 from every number

print(numbers.1 - 10)

# multiply 10 to every number

print(numbers.1 * 10)

# divide every number by 2

print(numbers.1 / 2)

mathematical operators 2

# perform addition on every member of numbers.1 and numbers.2

print(numbers.1 + numbers.2)

# perform subtraction on every member of numbers.1 and numbers.2

print(numbers.1 - numbers.2)

# perform division on every member of numbers.1 and numbers.2

print(numbers.1 / numbers.2)

# perform multiplication on every member of numbers.1 and

numbers.2
print(numbers.1 * numbers.2)

7 / 20
notes.md 10/3/2022

logical operators

numbers.3 = c(TRUE, TRUE, TRUE, FALSE, FALSE, TRUE, FALSE)

numbers.4 = c(FALSE, TRUE, TRUE, FALSE, FALSE, FALSE, TRUE)

# broadcast
# FALSE FALSE FALSE TRUE TRUE FALSE TRUE
print(!numbers.3)

# non-broadcast operators

# TRUE
print(numbers.3 && TRUE)

# TRUE
print(numbers.4 || TRUE)

# broadcast operators

# TRUE TRUE TRUE FALSE FALSE TRUE FALSE

print(numbers.3 & TRUE)

# TRUE TRUE TRUE TRUE TRUE TRUE TRUE

print(numbers.4 | TRUE)

Lists

multi-dimensional collection
collection made up with multiple vectors
to create a list call a function list()
e.g.
list created without temporary names

list.1 = list(
c(10, 20, 30, 40, 50),
c(60, 70, 80, 90, 100)
)

# [[1]]
# [1] 10, 20, 30, 40, 50
# [[2]]
# [2] 60, 70, 80, 90, 100

8 / 20
notes.md 10/3/2022

# 10, 20, 30, 40, 50

print(list.1[[1]])

# 20
print(list.1[[1]][2])

# 60, 70, 80, 90, 100

print(list.1[[2]])

# 90
print(list.1[[2]][4])

list created with temporary names

persons = list(
personNames = c("person1", "person2"),
personAddresses = c("pune", "mumbai")
)

# $personNames
# [1] "person1" "person2"
# #personAddresses
# [1] "pune" "mumbai"

# "person1" "person2"
print(persons$personNames)

# person2
print(persons$personNames[2])

# "pune" "mumbai"
print(persons$personAddresses)

# pune
print(persons$personAddresses[1])

Matrices

two dimensional collection

to create a matrix call a function named matrix()
e.g.

m.1 = matrix(c(10, 20, 30, 40), nrow = 2, ncol = 2)

9 / 20
notes.md 10/3/2022

# [, 1] [, 2]
# [1, ] 10 30
# [2, ] 20 40

m.2 = matrix(c(10, 20, 30, 40), nrow = 2, ncol = 2, byrow = TRUE)

# [, 1] [, 2]
# [1, ] 10 20
# [2, ] 30 40

retrieving values from matrix

m.1 = matrix(c(10, 20, 30, 40), nrow = 2, ncol = 2)

# [, 1] [, 2]
# [1, ] 10 30
# [2, ] 20 40

# [1] 10 30
print(m.1[1, ])

# [1] 30 40
print(m.1[, 2])

# [1] 30
print(m.1[1, 2])

to make the retrieval easier, matrix provides a naming rows and columns convention
e.g.

rownames = c("car1", "car2")

colnames = c("model", "average", "price")
cars = matrix(c("i20", "i10", 17, 18, 7.5, 5.5),
nrow = 2, ncol = 3, dimnames = list(rownames, colnames))

# "model" "average" "price"

# "car1" "i20" 17 7.5
# "car2" "i10" 18 5.5

# 7.5
print(cars["car1", "price"])

# i10
print(cars["car2", "model"])

10 / 20
notes.md 10/3/2022

recycling values while creating a matrix

m.5 = matrix(c(10, 20, 30), nrow = 3, ncol = 2)

# [,1] [,2]
# [1,] 10 10
# [2,] 20 20
# [3,] 30 30

m.6 = matrix(c(10, 20, 30), nrow = 3, ncol = 2, byrow = TRUE)

# [,1] [,2]
# [1,] 10 20
# [2,] 30 10
# [3,] 20 30

Arrays

Factors

the way to store categorical data

the values will get stored as numeric positions of levels
e.g.

factor.1 = factor(c('red', 'green', 'red', 'green', 'green'))

# red green red green green

# Levels: red green

factor.2 = factor(
c('red', 'green', 'red', 'green', 'green'),
levels = c('red', 'green'),
labels = c(1, 2)
)

# 1 2 1 2 2
# Levels: 1 2

Data Frames

tabular representation of the data

similar to pandas data frame
11 / 20
notes.md 10/3/2022

collection of rows and columns

two dimensional collection (Vectors)
list of lists
e.g.

names = c("person1", "person2", "person3", "person4")

ages = c(10, 20, 15, 18)
emails = c("[email protected]", "[email protected]", "[email protected]", "[email protected]")
addresses = c("pune", "mumbai", "nashik", "satara")

df.persons = data.frame(
name = names,
age = ages,
email = emails,
address = addresses
)

# name age email address

# 1 person1 10 [email protected] pune
# 2 person2 20 [email protected] mumbai
# 3 person3 15 [email protected] nashik
# 4 person4 18 [email protected] satara

functions
class
used to get the type of data structure
e.g.

v = c(10, 20, 30, 40, 50)

# numeric
print (class(v))

typeof
used to get the data type of data structure
e.g.

v = c(10, 20, 30, 40, 50)

12 / 20
notes.md 10/3/2022

# double
print (typeof(v))

names
used to set or get the temporary names for every position in the data structure
used with data.frames and lists
e.g.

list.1 = list(names = c("person1", "person2"), ages = c(20, 30))

# names ages
print(names(list.1))

str
similar to info() in pandas
used to get the basic information about the data structure
used with data frames and lists
the information includes
number of observations (rows)
number of variables (columns)
name of every variable
data type of every variable
preview of data in every column
e.g.

names = c("person1", "person2", "person3", "person4")

ages = c(10, 20, 15, 18)
emails = c("[email protected]", "[email protected]", "[email protected]",
"[email protected]")
addresses = c("pune", "mumbai", "nashik", "satara")

df.persons = data.frame(
name = names,
age = ages,
email = emails,
address = addresses
)

print(str(df.persons))

# 'data.frame': 4 obs. of 4 variables:

# $ name : chr "person1" "person2" "person3" "person4"
# $ age : num 10 20 15 18
# $ email : chr "[email protected]" "[email protected]" "[email protected]"
13 / 20
notes.md 10/3/2022

"[email protected]"
# $ address: chr "pune" "mumbai" "nashik" "satara"

summary
similar to describe() function in pandas
used to get statistical information about the data structure
can be used with data frames and lists
information includes
data type of column
for numeric column
minimum value
maxivalue
median
mean
1st quartile
3rd quartile
e.g.

names = c("person1", "person2", "person3", "person4")

ages = c(10, 20, 15, 18)
emails = c("[email protected]", "[email protected]", "[email protected]",
"[email protected]")
addresses = c("pune", "mumbai", "nashik", "satara")

df.persons = data.frame(
name = names,
age = ages,
email = emails,
address = addresses
)

print(summary(df.persons))

# name age email

address
# Length:4 Min. :10.00 Length:4
Length:4
# Class :character 1st Qu.:13.75 Class :character Class
:character
# Mode :character Median :16.50 Mode :character Mode
:character
# Mean :15.75
# 3rd Qu.:18.50
# Max. :20.00

head
14 / 20
notes.md 10/3/2022

used to retrieve first few records

e.g.

names = c("person1", "person2", "person3", "person4")

ages = c(10, 20, 15, 18)
emails = c("[email protected]", "[email protected]", "[email protected]",
"[email protected]")
addresses = c("pune", "mumbai", "nashik", "satara")

df.persons = data.frame(
name = names,
age = ages,
email = emails,
address = addresses
)

print(head(df.persons, 2))

# name age email address

# 1 person1 10 [email protected] pune
# 2 person2 20 [email protected] mumbai

tail
used to retrieve last few records
e.g.

names = c("person1", "person2", "person3", "person4")

ages = c(10, 20, 15, 18)
emails = c("[email protected]", "[email protected]", "[email protected]",
"[email protected]")
addresses = c("pune", "mumbai", "nashik", "satara")

df.persons = data.frame(
name = names,
age = ages,
email = emails,
address = addresses
)

print(tail(df.persons, 2))

# name age email address

# 3 person3 15 [email protected] nashik
# 4 person4 18 [email protected] satara

nrow
15 / 20
notes.md 10/3/2022

used to get number of rows in a data frame

e.g.

names = c("person1", "person2", "person3", "person4")

ages = c(10, 20, 15, 18)
emails = c("[email protected]", "[email protected]", "[email protected]",
"[email protected]")
addresses = c("pune", "mumbai", "nashik", "satara")

df.persons = data.frame(
name = names,
age = ages,
email = emails,
address = addresses
)

# 4
print(nrow(df))

ncol
used to get number of columns in a data frame
e.g.

names = c("person1", "person2", "person3", "person4")

ages = c(10, 20, 15, 18)
emails = c("[email protected]", "[email protected]", "[email protected]",
"[email protected]")
addresses = c("pune", "mumbai", "nashik", "satara")

df.persons = data.frame(
name = names,
age = ages,
email = emails,
address = addresses
)

# 4
print(ncol(df))

colnames
used to get column names
e.g.

16 / 20
notes.md 10/3/2022

names = c("person1", "person2", "person3", "person4")

ages = c(10, 20, 15, 18)
emails = c("[email protected]", "[email protected]", "[email protected]",
"[email protected]")
addresses = c("pune", "mumbai", "nashik", "satara")

df.persons = data.frame(
name = names,
age = ages,
email = emails,
address = addresses
)

# "name" "age" "email" "address"

print(colnames(df))

rownames
used to get row names
e.g.

names = c("person1", "person2", "person3", "person4")

ages = c(10, 20, 15, 18)
emails = c("[email protected]", "[email protected]", "[email protected]",
"[email protected]")
addresses = c("pune", "mumbai", "nashik", "satara")

df.persons = data.frame(
name = names,
age = ages,
email = emails,
address = addresses
)

# "1" "2" "3" "4"

print(rownames(df))

Type inspection
in R type can be inspected by using is functions (functions start with is.*)
e.g.

num = 100
print(is.numeric(num)) # TRUE

17 / 20
notes.md 10/3/2022

print(is.character(num)) # FALSE

first.name = "steve"
print(is.numeric(first.name)) # FALSE
print(is.character(first.name)) # TRUE

Type conversion
used to convert data from one type to another
e.g.

print(as.numeric("10")) # 10

print(as.logical("TRUE")) # TRUE

print(as.character(10)) # "10"
print(as.character(FALSE)) # "FALSE"

print(as.logical(1)) # TRUE
print(as.logical(0)) # FALSE
print(as.logical(100)) # TRUE
print(as.logical(-100)) # TRUE

loops
for loop

numbers.1 = c(10, 20, 30, 40, 50)

for (value in numbers.1) {

print(value)
}

functions

custom function
user defind functions
fucntion can be created with function object
syntax

18 / 20
notes.md 10/3/2022

# function declaration

# empty function
function1 = function() {
# function body here
}

# function call
function1()

built-in fuction
getwd
used to get the current working directory
working directory is used to search the files/documents when application wants to read
them
setwd
used to set the current working directory
working directory is used to search the files/documents when application wants to read
them
e.g.

setwd('/Volumes/data/sunbeam/dataset/data1/')

install.package
used to install a package or library and its dependencies
e.g.

install.package('tidyverse')

library
used to load a library in the current environment
e.g.

library(tidyverse)

19 / 20
notes.md 10/3/2022

version numbers
version number: x.y.z.a
x: major version
y: minor version
z: build version
a: nightly build version
change in the version number
major number: breaking changes
minor number: enhancement / feature addition / bug fixing
build number: bug fixing / nightly build

20 / 20