Freq Out!

Overview

Load Libraries

options(scipen = 999)
library(tidyverse)

Registered S3 methods overwritten by 'dbplyr':
  method         from
  print.tbl_lazy     
  print.tbl_sql      
-- Attaching packages --------------------------------------------------------------------------------------------------------------------- tidyverse 1.3.0 --
v ggplot2 3.3.3     v purrr   0.3.4
v tibble  3.0.4     v dplyr   1.0.2
v tidyr   1.1.2     v stringr 1.4.0
v readr   1.4.0     v forcats 0.5.0
-- Conflicts ------------------------------------------------------------------------------------------------------------------------ tidyverse_conflicts() --
x dplyr::filter() masks stats::filter()
x dplyr::lag()    masks stats::lag()

Load data

boston  <- read_csv("boston_2020.csv")

-- Column specification --------------------------------------------------------------------------------------------------------------------------------------
cols(
  .default = col_character(),
  PID = col_double(),
  AV_TOTAL = col_double(),
  LAND_SF = col_double(),
  YR_BUILT = col_double(),
  YR_REMOD = col_double(),
  LIVING_AREA = col_double(),
  NUM_FLOORS = col_double(),
  STRUCTURE_CLASS = col_logical(),
  R_TOTAL_RMS = col_double(),
  R_BDRMS = col_double(),
  R_FULL_BTH = col_double(),
  R_HALF_BTH = col_double(),
  R_FPLACE = col_double()
)
i Use `spec()` for the full column specifications.

zips    <- read_csv("boston_zips.csv")

-- Column specification --------------------------------------------------------------------------------------------------------------------------------------
cols(
  ZIP = col_character(),
  Population = col_double(),
  Pop_Density = col_double(),
  Median_Income = col_double(),
  City_State = col_character()
)

truecar <- read_csv("true_car_prices_50k.csv")

-- Column specification --------------------------------------------------------------------------------------------------------------------------------------
cols(
  vin = col_character(),
  make = col_character(),
  model = col_character(),
  year = col_double(),
  price = col_double(),
  mileage = col_double(),
  city = col_character(),
  state = col_character(),
  region = col_character(),
  population = col_double(),
  lat = col_double(),
  lng = col_double()
)

Top N Frequency

framework for our function, here we want to create working code of what we want our function to do.

1. we want to create a top 20 frequency per character column
2. we want to plot the frequency

freq <- truecar %>%
  group_by(make) %>%
  summarise(n = n()) %>%
  mutate(pct = n/sum(n)) %>%
  arrange(desc(n)) %>%
  top_n(20,n)

`summarise()` ungrouping output (override with `.groups` argument)

print(freq)

freq %>%
  ggplot(aes(reorder(make,n),n)) +
  geom_col() +
  coord_flip() +
  labs(title = "Frequency Analysis",
       y = "count",
       x = "category")

Next wrap code in a function

like this.

freq_function <- function(){
  # frequency analysis of true car by make 
  freq <- truecar %>%
    group_by(make) %>%
    summarise(n = n()) %>%
    mutate(pct = n/sum(n)) %>%
    arrange(desc(n)) %>%
    top_n(20,n)
  
  print(freq)
  
  freq %>%
    ggplot(aes(reorder(make,n),n)) +
    geom_col() +
    coord_flip() +
    labs(title = "Frequency Analysis",
         y = "count",
         x = "category")
}

freq_function()

Parameterize your function

add the argument data and replace truecar with the argument

freq_function <- function(data){
  # frequency analysis of true car by make 
  freq <- data %>%
    group_by(make) %>%
    summarise(n = n()) %>%
    mutate(pct = n/sum(n)) %>%
    arrange(desc(n)) %>%
    top_n(20,n)
  
  print(freq)
  
  freq %>%
    ggplot(aes(reorder(make,n),n)) +
    geom_col() +
    coord_flip() +
    labs(title = "Frequency Analysis",
         y = "count",
         x = "category")
}

freq_function(truecar)

`summarise()` ungrouping output (override with `.groups` argument)

Next replace column

you want to pass it a column to perform frequency analysis of to do this you need to tell R that the string passed is in fact a column. to do this we use !!as.name(“string”) this tells R that the string is a column name

1. add an argument column to your function
2. repalce make with !!as.name(column)
3. test with make, model, year, city, state

freq_function <- function(data, column){
  # frequency analysis of true car by make 
  freq <- data %>%
    group_by(!!as.name(column)) %>%
    summarise(n = n()) %>%
    mutate(pct = n/sum(n)) %>%
    arrange(desc(n)) %>%
    top_n(20,n)
  
  print(freq)
  
  freq %>%
    ggplot(aes(reorder(!!as.name(column),n),n)) +
    geom_col() +
    coord_flip() +
    labs(title = "Frequency Analysis",
         subtitle = column,
         y = "count",
         x = "category")
}
freq_function(truecar, "make")

`summarise()` ungrouping output (override with `.groups` argument)

freq_function(truecar, "model")

`summarise()` ungrouping output (override with `.groups` argument)

freq_function(truecar, "year")

`summarise()` ungrouping output (override with `.groups` argument)

freq_function(truecar, "city")

`summarise()` ungrouping output (override with `.groups` argument)

freq_function(truecar, "state")

`summarise()` ungrouping output (override with `.groups` argument)

Function 1 GET_STATS

get_stats(dataframe, column)

your challenge is to add the mean, min, max, and column name to the table returned by the function call.

get_stats <- function(dataframe, column){
  dataframe %>%
    summarise(n = n(),
              n_distinct = n_distinct(!!as.name(column)),
              n_miss = sum(is.na(!!as.name(column))),
              mean = mean(!!as.name(column)),
              min = min(!!as.name(column)),
              max = max(!!as.name(column))) %>%
    mutate( column = column)
  
              # add mean, min, max 
               # %>%
    # add a column name ~ mutate(column = column)
}

get_stats(boston,"AV_TOTAL")

get_stats(boston, "LIVING_AREA")

get_stats(truecar, "price")

get_stats(truecar, "mileage")

Function 2 get_chart(dataframe, group_by_column, mean_column)

group_by_column <- "ZIPCODE"
mean_column <- "AV_TOTAL"
dataset <- boston 

get_chart <- function(dataframe, group_by_column, mean_column){
  res <- dataframe %>%
    group_by(!!as.name(group_by_column)) %>%
    summarise(mean = mean(!!as.name(mean_column))) # change this 
  
  res %>% ggplot(aes(reorder(!!as.name(group_by_column), mean), mean)) +
    geom_col() +
    labs(title = paste("mean ",mean_column, " by ", group_by_column),
         x = "ZIPCODE", 
         y = "Mean AV_TOTAL", # fix this 
         caption = "Dataset: Boston")
}

group_by_column2 <- "R_BLDG_STYL"
mean_column2 <- "LIVING_AREA"
dataset <- boston 

get_chart2 <- function(dataframe, group_by_column2, mean_column2){
  res <- dataframe %>%
    group_by(!!as.name(group_by_column2)) %>%
    summarise(mean = mean(!!as.name(mean_column2))) # change this 
  
  res %>% ggplot(aes(reorder(!!as.name(group_by_column2), mean), mean)) +
    geom_col() +
    theme(axis.text.x = element_text(angle = 45, hjust = 1)) +
    labs(title = paste("mean ",mean_column2, " by ", group_by_column2),
         x = "R_BLDG_STYL", 
         y = "Mean LIVING_AREA", # fix this 
         caption = "Dataset: Boston")
}

group_by_column3 <- "make"
mean_column3 <- "price"
dataset <- truecar 

get_chart3 <- function(dataframe, group_by_column3, mean_column3){
  res <- dataframe %>%
    group_by(!!as.name(group_by_column3)) %>%
    summarise(mean = mean(!!as.name(mean_column3))) # change this 
  
  res %>% ggplot(aes(reorder(!!as.name(group_by_column3), mean), mean)) +
    geom_col() +
    theme(axis.text.x = element_text(angle = 45, hjust = 1)) +
    labs(title = paste("mean ",mean_column3, " by ", group_by_column3),
         x = "make", 
         y = "Mean price", # fix this 
         caption = "Dataset: truecar")
}

group_by_column4 <- "make"
mean_column4 <- "mileage"
dataset <- truecar 

get_chart4 <- function(dataframe, group_by_column4, mean_column4){
  res <- dataframe %>%
    group_by(!!as.name(group_by_column4)) %>%
    summarise(mean = mean(!!as.name(mean_column4))) # change this 
  
  res %>% ggplot(aes(reorder(!!as.name(group_by_column4), mean), mean)) +
    geom_col() +
    theme(axis.text.x = element_text(angle = 45, hjust = 1)) +
    labs(title = paste("mean ",mean_column4, " by ", group_by_column4),
         x = "make", 
         y = "Mean mileage", # fix this 
         caption = "Dataset: truecar")
}

get_chart(boston, "ZIPCODE", "AV_TOTAL")

`summarise()` ungrouping output (override with `.groups` argument)

get_chart2(boston, "R_BLDG_STYL", "LIVING_AREA")

`summarise()` ungrouping output (override with `.groups` argument)

get_chart3(truecar, "make", "price")

`summarise()` ungrouping output (override with `.groups` argument)

get_chart4(truecar, "make", "mileage")

`summarise()` ungrouping output (override with `.groups` argument)