Unverified Commit 4a53e20d by Arham Akheel Committed by GitHub

Code changes

parent 895a1f55
################################################################################ ################################################################################
## This code is property of Data Science Dojo ## This code is property of Data Science Dojo
## Copyright (C) 2017~2018 ## Copyright (C) 2017~2018
## ##
## Objective: Manipulate and visualize data using R ## Objective: Manipulate and visualize data using R
## Please install "dplyr" package: install.packages("dplyr") ## Please install "dplyr" package: install.packages("dplyr")
## Please install "ggplot2" package: install.packages("ggplot2") ## Please install "ggplot2" package: install.packages("ggplot2")
################################################################################ ################################################################################
# Script for following along in Introduction to dplyr # Script for following along in Introduction to dplyr
# Copy-paste line by line or use the "Run" button in R Studio # Copy-paste line by line or use the "Run" button in R Studio
#Set the working directory, example: setwd("directory/dataset folder") #Set the working directory, example: setwd("directory/dataset folder")
install.packages("dplyr") install.packages("dplyr")
install.packages("ggplot2") install.packages("ggplot2")
library(dplyr) library(dplyr)
library(ggplot2) library(ggplot2)
setwd("C:/Users/Arham/Desktop/Files/Introduction to dplyr")
#Reading the dataset from the working directory. #Reading the dataset from the working directory.
#Setting string values as characters #Setting string values as characters
#loading the greek characters #loading the greek characters
wine = read.csv("wine.csv", stringsAsFactors = FALSE, encoding = 'UTF-8') wine = read.csv("wine.csv",
stringsAsFactors = FALSE,
View (wine) encoding = 'UTF-8')
#Removing columns from dataset View (wine)
wine = wine[,-c(1,3)]
#Removing columns from dataset
#Creating a dataset by counting all observations grouped by country and then creating a new variable called count wine = wine[,-c(1,3)]
wine %>% group_by(country) %>% summarize(count=n()) %>% arrange(desc(count))
#Creating a dataset by counting all observations grouped by country and then creating a new variable called count
#Creating a new variable which contains the top 10 countries wine %>%
selected_countries = wine %>% group_by(country) %>% summarize(count=n()) %>% arrange(desc(count)) %>% top_n(10) %>% select(country) group_by(country)%>%
summarize(count=n()) %>%
selected_countries arrange(desc(count))
#Creating a new variable which contains the top 10 countries
#Changing the format from data frame to vector as.character referencing the country column selected_countries = wine %>%
selected_countries = as.character(selected_countries$country) group_by(country) %>%
class(selected_countries) summarize(count=n()) %>%
arrange(desc(count)) %>%
#Subsetting data selecting top ten countries and their points from wine top_n(10) %>%
select_points=wine %>% filter(country %in% selected_countries) %>% select(country, points) %>% arrange(country) select(country)
#Scatterplot with smooth line selected_countries
ggplot(wine, aes(points,price)) + geom_point() + geom_smooth()
#Boxplot between country and points, reordered by median of points. Center aligning the Title of the boxplot #Changing the format from data frame to vector as.character referencing the country column
ggplot(select_points, aes(x=reorder(country,points,median),y=points)) + geom_boxplot(aes(fill=country)) + xlab("Country") + ylab("Points") + ggtitle("Distribution of Top 10 Wine Producing Countries") + theme(plot.title = element_text(hjust = 0.5)) selected_countries = as.character(selected_countries$country)
#Filter by countries that do not appear on the selected_countries dataset class(selected_countries)
wine %>% filter(!(country %in% selected_countries)) %>% group_by(country) %>% summarize(median=median(points)) %>% arrange(desc(median))
#Subsetting data selecting top ten countries and their points from wine
#Creating a new variable called top using country and points to rate them based on points select_points=wine %>%
top=wine %>% group_by(country) %>% summarize(median=median(points)) %>% arrange(desc(median)) filter(country %in% selected_countries) %>%
class(top) select(country, points) %>%
arrange(country)
#Changing the format from data frame to vector as.character referencing the country column
top=as.character(top$country) #Scatterplot with smooth line
top ggplot(wine, aes(points,price)) +
geom_point() +
#Using intersect function to select the common values in both datasets geom_smooth()
both=intersect(top,selected_countries)
both #Boxplot between country and points, reordered by median of points. Center aligning the Title of the boxplot
ggplot(select_points,
#Using setdiff to select the non-overlapping values in both datasets aes(x=reorder(country,points,median),
not = setdiff(top, selected_countries) y=points)) +
not geom_boxplot(aes(fill=country)) +
xlab("Country") +
#Creating a subset based on variety using group by and summarize ylab("Points") +
topwine = wine %>% group_by(variety) %>% summarize(number=n()) %>% arrange(desc(number)) %>% top_n(10) ggtitle("Distribution of Top 10 Wine Producing Countries") +
topwine=as.character(topwine$variety) theme(plot.title = element_text(hjust = 0.5))
topwine
#Filter by countries that do not appear on the selected_countries dataset
#Plot based on variety and points using group by and summarize wine %>%
wine %>% filter(variety %in% topwine) %>% group_by(variety)%>% summarize(median=median(points)) %>% ggplot(aes(reorder(variety,median),median)) + geom_col(aes(fill=variety)) + xlab('Variety') + ylab('Median Point') + scale_x_discrete(labels=abbreviate) filter(!(country %in% selected_countries)) %>%
group_by(country) %>%
#Creating top 15 percent cheapest wines with high rating using intersect function summarize(median=median(points)) %>%
top15percent=wine %>% arrange(desc(points)) %>% filter(points > quantile(points, prob = 0.85)) arrange(desc(median))
cheapest15percent=wine %>% arrange(price) %>% head(nrow(top15percent))
goodvalue = intersect(top15percent,cheapest15percent) #Creating a new variable called top using country and points to rate them based on points
goodvalue top=wine %>%
group_by(country) %>%
#Feature Engineering summarize(median=median(points)) %>%
arrange(desc(median))
wine = read.csv('wine.csv', stringsAsFactors = FALSE, encoding = 'UTF-8')
class(top)
save(wine, file = "wine.rda")
load("wine.rda") #Changing the format from data frame to vector as.character referencing the country column
top=as.character(top$country)
#Omiting one column from the wine dataset top
wine = wine[,-c(3)]
#Using intersect function to select the common values in both datasets
View(wine) both=intersect(top,selected_countries)
both
#Using transmute and mutate functions to append a new column
wine1 = wine %>% mutate(PPratio = points/price) #Using setdiff to select the non-overlapping values in both datasets
wine2 = wine %>% transmute(PPratio = points/price) not = setdiff(top, selected_countries)
not
#Aggregation by country using group by and summarize
wine %>% group_by(country) %>% summarize(total = n()) #Creating a subset based on variety using group by and summarize
topwine = wine %>%
#Missing country values group_by(variety) %>%
wine[wine$country == "",] summarize(number=n()) %>%
arrange(desc(number)) %>%
#Adding missing values in the dataset top_n(10)
wine$country = ifelse(wine$designation == "Askitikos", "Greece", wine$country)
wine$country = ifelse(wine$designation == "Piedra Feliz", "Chile", wine$country) topwine=as.character(topwine$variety)
wine$country = ifelse(wine$variety == "Red Blend", "Turkey", wine$country)
topwine
#Combining Datasets
#Creating a new subset by total number of rows by country #Plot based on variety and points using group by and summarize
newwine = wine %>% group_by(country) %>% summarize(total = n()) %>% arrange(desc(total)) wine %>%
filter(variety %in% topwine) %>%
#Creating subsets with the head of wine and newwine group_by(variety)%>%
subset1=head(wine) summarize(median=median(points)) %>%
subset2=head(newwine) ggplot(aes(reorder(variety,median),median)) +
geom_col(aes(fill=variety)) +
#Combining two data frames using full join function xlab('Variety') + ylab('Median Point') +
full = full_join(subset1, subset2) scale_x_discrete(labels=abbreviate)
full
#Creating top 15 percent cheapest wines with high rating using intersect function
#Combining two data frames using inner join function top15percent=wine %>%
inner = inner_join(subset1, subset2) arrange(desc(points)) %>%
inner filter(points > quantile(points, prob = 0.85))
#Combining two data frames using left join function cheapest15percent=wine %>%
left = left_join(subset1, subset2) arrange(price) %>%
left head(nrow(top15percent))
#Combining two data frames using right join function goodvalue = intersect(top15percent,cheapest15percent)
right = right_join(subset1, subset2)
right goodvalue
#####End of Code#### #Feature Engineering
wine = read.csv('wine.csv',
stringsAsFactors = FALSE,
encoding = 'UTF-8')
save(wine, file = "wine.rda")
load("wine.rda")
#Omiting one column from the wine dataset
wine = wine[,-c(3)]
View(wine)
#Using transmute and mutate functions to append a new column
wine1 = wine %>%
mutate(PPratio = points/price)
wine2 = wine %>%
transmute(PPratio = points/price)
#Aggregation by country using group by and summarize
wine %>%
group_by(country) %>%
summarize(total = n())
#Missing country values
wine[wine$country == "",]
#Adding missing values in the dataset
wine$country =
ifelse(wine$designation == "Askitikos",
"Greece", wine$country)
wine$country =
ifelse(wine$designation == "Piedra Feliz",
"Chile", wine$country)
wine$country =
ifelse(wine$variety == "Red Blend",
"Turkey", wine$country)
#Combining Datasets
#Creating a new subset by total number of rows by country
newwine = wine %>%
group_by(country) %>%
summarize(total = n()) %>%
arrange(desc(total))
#Creating subsets with the head of wine and newwine
subset1=head(wine)
subset2=head(newwine)
#Combining two data frames using full join function
full = full_join(subset1, subset2)
full
#Combining two data frames using inner join function
inner = inner_join(subset1, subset2)
inner
#Combining two data frames using left join function
left = left_join(subset1, subset2)
left
#Combining two data frames using right join function
right = right_join(subset1, subset2)
right
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment