Master R 4 - Restruct Data

rm(list=ls())

# Transposing matrices
m <- matrix(1:9, 3)
t(m)

# Filtering data by string matching
library(plyr)
library(dplyr)
library(hflights)

str(select(hflights, ends_with("delay")))
str(select(hflights, contains("T", ignore.case = FALSE)))
str(select(hflights, matches("^[[:alpha:]]{5,6}$")))
table(nchar(names(hflights)))
str(select(hflights, -matches("^[[:alpha:]]{7,8}$")))

# Rearranging data
str(arrange(hflights, ActualElapsedTime))
install.packages("magrittr")
library(magrittr)
hflights %>% arrange(ActualElapsedTime) %>% str
hflights %>% arrange(ActualElapsedTime) %>%
  select(ActualElapsedTime, Dest) %>%
  subset(Dest != 'AUS') %>%
  head %>%
  str
library(data.table)
str(head(data.table(hflights, key='ActualElapsedTime')[Dest != 'AUS', c('ActualElapsedTime', 'Dest'), with = FALSE]))
str(head(na.omit(data.table(hflights, key='ActualElapsedTime'))[Dest != 'AUS', list('ActualElapsedTime', 'Dest')]))

# dplyr vs data.table

# Computing new varibles
hflights_dt <- data.table(hflights)
hflights_dt[, DistanceKMs := Distance / 0.62137]
hflights_dt$DistanceKMs <- hflights_dt$Distance / 0.62137

# Memory profiling
install.packages("pryr")
library(pryr)
hflights_dt <- data.table(hflights)
address(hflights_dt)
hflights_dt$DistanceKMs <- hflights_dt$Distance / 0.62137

hflights_dt <- data.table(hflights)
address(hflights_dt)
hflights_dt[, DistanceKMs := Distance / 0.62137]
address(hflights_dt)
system.time(within(hflights_dt, DistanceKMs <- Distance / 0.62137))

# Creating multiple variables at a time
hflights_dt[, c('DistanceKMs', 'DistanceFeets') := list(Distance/0.62137, Distance * 5280)];
carriers <- unique(hflights_dt$UniqueCarrier)
hflights_dt[, paste('carrier', carriers, sep = '_') := lapply(carriers, function(x) as.numeric(UniqueCarrier == x))]
str(hflights_dt[, grep('^carrier', names(hflights_dt)), with = FALSE])
str(hflights_dt)

# Computing new variables with dplyr
hflights <- hflights %>% mutate(DistanceKMs = Distance / 0.62137)
hflights <- mutate(hflights, DistanceKMs = Distance/0.62137)

# Merging datasets
wdays <- data.frame(DayOfWeek = 1: 7,DayOfWeekString = c("Sunday", "Monday", "Tuesday","Wednesday", "Thursday", "Friday", "Saturday"))
system.time(merge(hflights, wdays))
system.time(merge(hflights_dt, wdays, by = 'DayOfWeek'))
hflights$wdays <- weekdays(as.Date(with(hflights, paste(Year, Month, DayofMonth, sep = '-'))))

# One of the most often used function along with the base commands, rbind and cbind is do.call, which can execute the rbind or cbind call on all elemetns of a list, thus enabling us, to join a list of data frames.
# Similarly, rbindlist can be called to merge a list of data.table objects ina much faster way.

# Reshaping data in a flexible way

# Converting wide tables to the long table format
library(reshape2)
head(melt(hflights))
hflights_melted <- melt(hflights, id.vars=0, measure.vars = c('ActualElapsedTime', 'AirTime'))
str(hflights_melted)

library(ggplot2)
ggplot(hflights_melted, aes(variable, y=value)) + geom_boxplot()

# Converting long tables to the wide table format
hflights_melted <- melt(hflights, id.vars='Month', measure.vars = c('ActualElapsedTime', 'AirTime'))
df <- dcast(hflights_melted, Month ~ variable, fun.aggregate = mean, na.rm = TRUE)
ggplot(melt(df, id.vars = 'Month')) +
  geom_line(aes(x=Month, y=value, color=variable)) +
  scale_x_continuous(breaks = 1:12) +
  theme_bw() +
  theme(legend.position = 'top')

hflights_melted <- melt(add_margins(hflights, 'Month'), id.vars = 'Month', measure.vars = c('ActualElapsedTime', 'AirTime'))
df <- dcast(hflights_melted, Month ~ variable, fun.aggregate = mean, na.rm = TRUE)