Learning-R

Learning R Language

Importing libraries:

library(dplyr)

Importing data from packages:

data(gapminder)

Variable Types:

numeric: int and floats
character: char or strings
logical: booleans

Assigning values to a variable:

apples <- 5
oranges <- 4
my_string <- "Teste R"

Operations

fruits <- apples + oranges

Booleans

logica <- TRUE

Functions to retrieve info about the dataframes / variables

str(variable)
glimpse(dataset)

Class Method

'class' returns the variable type

class(my_numeric)

UNIQUE

unique(cars$ncyl)

Vectors

vectors are unordered and have the same type

if you add one single char type value, all the elements of the vector turn to char type

vetor <- c(1, 2, 0, -5, 100)

Assigning labels to the vector's elements using 'names' method:

names(vetor) = c("Monday", "Tuesday", "Wednesday", "Thursday", "Friday")

You can also create a label vector to simplify atributtion and create 'dictionaries':

nomes <- c("Monday", "Tuesday", "Wednesday", "Thursday", "Friday")
names(vetor) <- nomes

Operation with vectors

vetor1 <- c(1, 2, 3, 4, 5)
vetor2 <- c(0, 1, 3, -1, 10)
vetor_soma <- vetor1 + vetor2

SUM method

vetor2_soma <- sum(vetor2)
vetor_soma <- sum(vetor1, vetor2)

Accessing elements from a vector

In R, the first element is reffered as position 1, not 0 as in python

vetor3 <- vetor2[3]
vetor4 <- vetor1[c(1:3)]
poker_midweek <- poker_vector[c(2:4)]

Accessing through element label:

vetor3 <- vetor["Monday"]

Comparison: returns booleans

vetor_comparison <- vetor > 2

Selecting values corresponding to the boolean vector

vetor_comparado <- vetor[vetor_comparison]

MEAN method:

media <- mean(vetor)

MATRIX

Fills a matrix 3x3 with range 1:9 by row

matriz = matrix(1:9, nrow = 3, byrow = TRUE)

Fills a matrix with vector's elements

star_wars_matrix <- matrix(box_office, nrow = 3, byrow = TRUE)

FACTOR

describes categorical variables (ex: sex, color, etc)

factor method defines levels (i.e., the unique cathegories:

survey_vector <- c("M", "F", "F", "M", "M")
factor_survey_vector <- factor(survey_vector)

Apply the factor function with required order of the level.

new_order_data <- factor(factor_data,levels = c("East","West","North"))

Specify the levels of factor_survey_vector

Levels may also be used to exhibit the unique cathegories of a variable (ex: levels(sex) >> male, female )

levels(factor_survey_vector) <- c("Female", "Male")

ordering the cathegory levels

temperature_vector <- c("High", "Low", "High","Low", "Medium")
factor_temperature_vector <- factor(temperature_vector, order = TRUE, levels = c("Low", "Medium", "High"))

DATASETS

head(dataset) : returns the first lines

tail(dataset): returns the last lines

str(dataset): checks the structure of the dataframe

data.frame(vetor1, vetor2, vetor3, vetorn...) creates a dataframe from one or multiple vectors

subset select data according to specified conditions

subset(df, subset = column1 > 1)

LISTS

Lists can be constructed by list() function, with any elements, such as strings, dataframes, vectors, matrixes, numbers, etc.

lista <- list("a", "b", 1, 7, "dsfdsdf")
lista2 <- list(1, "a", vetor1, df, matriz1)

labeling list's elements

lista <- list(vec = vetor1, mat = matriz1, df=my_df)

selecting elements

select the 2nd element from a vector 'vetor1' in the 3rd position of the list 'lista2'

lista2 [[3]] [2] or
lista2$vetor1[2] or

adding new elements with name to the list

lista1 <- c(lista2, ano=1980)

FILTER

filter(people, hair_color == "none", eye_color == "black")

to filter by discrete values

filter(cars, ncyl %in% c(4, 6, 8))

filtering a numerical interval

filter(MedGPA, GPA >= 3.375 & GPA <= 3.77)

PIPE WITH FILTER

use filter to select values from a column, pipe this selection into a dataset and then apply to a new dataset

filtered_dataset <- dataset %>%
filter(coluna == "blue")

MUTATE

change values iteratively according to a condition

ifelse works as excel's 'if' function

mutate(num_char_cat = ifelse(num_char < med_num_char, "below median", "at or above median"))

Filtering Asia countries, then adding column indicating outliers (life expect under 50)

gap_asia <- gap2007 %>%
filter(continent=="Asia") %>%
mutate(is_outlier = (lifeExp <50))

Creating a new variable 'noise' filled with values of rnorm() function and inserting into 'mario_kart' dataset

rnorm(N) generates a vector of N pseudo-random normals with mean 0 and variance 1. N had to be the number of observations (lines)

mario_kart_noisy <- mario_kart %>% mutate(noise = rnorm(141))

TABLE

May be used to create a 2-way table to summarize the data in two columns

table(df$a, df$b)

DROPLEVELS

drop all the columns that don't fill a specified condition (filter or factor)

comics <- comics %>%
filter(align != "Reformed Criminals") %>%
droplevels()

SUMMARIZE

apply function(s) in a dataframe

gap2007 %>%
group_by(continent) %>%
summarize(mean(lifeExp), median(lifeExp))