Introduction to R

Jeffery Russell
9-30-19

RIT Scientific Computing Group

History

Back in the day (1976) Bell Laboratories created the S statistical programming language
People were sad because it was exclusively licensed by AT&T
During the 90's a group of people developed a S replacement called R and it was licensed under GNU

Why use R

Statistics and data analysis
Machine learning
Fast prototyping
Creating graphs
Writing research papers and reports

R Markdown

Using R markdown you can easily create reports and presentations by embedding your code in the report. This has major advantages.

Easier to reproduce other people's research
Easy to re-run your analysis if the data changes – no more re-generating 10 charts and re-computing statistical tests
Easy to export to other formats

Embedding Code Output in a Document

summary(cars)

     speed           dist       
 Min.   : 4.0   Min.   :  2.00  
 1st Qu.:12.0   1st Qu.: 26.00  
 Median :15.0   Median : 36.00  
 Mean   :15.4   Mean   : 42.98  
 3rd Qu.:19.0   3rd Qu.: 56.00  
 Max.   :25.0   Max.   :120.00

Embedding Graphs in a Document

plot(mtcars$wt, mtcars$mpg, main="Weight vs MPG", xlab = "weight", ylab="MPG")

plot of chunk unnamed-chunk-2

Syntax

Syntax of R is C-esk with its use of curly braces
Variables are similar to python since it infers your data type used
Type system is rather weird, the base unit for everything is a vector– even integers

x <- 0
if (x < 0) 
{
  print("Negative number")
} else if (x > 0) 
{
  print("Positive number")
} else
{
  print("Zero")
}

[1] "Zero"

Syntax: For Loops

for(i in 1:5)
{
  print(i)
}

[1] 1
[1] 2
[1] 3
[1] 4
[1] 5

for(i in (1:5)*2)
{
  print(i)
}

[1] 2
[1] 4
[1] 6
[1] 8
[1] 10

Syntax: While Loops

x <- 2

while(x == 2)
{
  print("Stonks")
  x = x + 1
}

[1] "Stonks"

Syntax cont

Syntax: Arrays

Arrays are 1 indexed.

for(i in c(1,4,5))
{
  print(i)
}

[1] 1
[1] 4
[1] 5

ar <- c(1,3,9)

print(ar[2])

[1] 3

Getting Help

Using the built in help command, you can view documentation for any function.

help(plot)

Making Graphs

plot(x=1:10, y=(1:10)^2, xlab = "x", ylab = "y", main="Ez")

plot of chunk unnamed-chunk-10

ML Example pt: 1

ggplot(data = iris) + 
  theme(plot.title = element_text(hjust = 0.5)) + 
  ggtitle("Iris Flower Set") + 
  geom_point(mapping = aes(x=Sepal.Length, y=Petal.Length, color = Species)) + 
  labs(x = "Sepal Length", y = "Petal Length", color="Phase") +
  theme_bw()

plot of chunk unnamed-chunk-11

ML Example pt: 2

ggplot(data = iris) + theme(plot.title = element_text(hjust = 0.5)) + 
  ggtitle("Iris Flow Length") + 
  geom_boxplot(mapping = aes(y=Petal.Length, x = Species), outlier.colour = "red", outlier.shape = 1) + 
  labs(x = "Flower Type", y = "Petal Length") +
  coord_flip() +
  theme_bw()

plot of chunk unnamed-chunk-12

Super Cool ML Example pt: 3

sc <- spark_connect(master = "local")

iris_tbl <- sdf_copy_to(sc, iris, name = "iris_tbl", overwrite = TRUE)

partitions <- iris_tbl %>%
  sdf_partition(training = 0.7, test = 0.3, seed = 1111)

iris_training <- partitions$training
iris_test <- partitions$test

dt_model <- iris_training %>%
  ml_decision_tree(Species ~ .)

pred <- ml_predict(dt_model, iris_test)

ml_multiclass_classification_evaluator(pred)

[1] 0.9451737

Resources

Questions?

Workshop

Visualize the built in “mpg” data-set from the tidyverse library.

Step 1: Install Tidyverse package in R and include it in your R Script

# Install tidyverse
install.packages("tidyverse")

# Include tidyverse in project

library(tidyverse)

Engine Size Vs MPG Normal Plot

plot(x = mpg$displ, y=mpg$hwy, main="Engine Size(Liters) vs MPG")

plot of chunk unnamed-chunk-14

Engine Size Vs MPG

ggplot(data = mpg) + geom_point(mapping=aes(x=displ, y = hwy))

plot of chunk unnamed-chunk-15

Engine Size Vs MPG with color Mappings

ggplot(data = mpg) + geom_point(mapping=aes(x=displ, y = hwy, color=class))

plot of chunk unnamed-chunk-16

Engine Size Vs MPG with shape Mappings

ggplot(data = mpg) + geom_point(mapping=aes(x=displ, y = hwy, shape=class))

plot of chunk unnamed-chunk-17

Engine Size Vs MPG For each Class

ggplot(data = mpg) + geom_point(mapping=aes(x=displ, y = hwy)) + facet_wrap(~ class, nrow=3)

plot of chunk unnamed-chunk-18

Engine Size Vs MPG With Line of Best Fit

ggplot(data = mpg) + 
  geom_point(aes(x=displ, y = hwy, color=class))+
  geom_smooth(aes(x=displ, y = hwy))

plot of chunk unnamed-chunk-19

Engine Size vs MPG With Line of Best Fit

ggplot(data = mpg, aes(x=displ, y = hwy, color=class)) + 
  geom_point()+
  geom_smooth()

plot of chunk unnamed-chunk-20