yaojenkuo

def write_concatenated_df():
    url_2011 = "https://www.run2pix.com/report/report_w.php?EventCode=20111218&Race=MA&sn=3"
    url_2012 = "https://www.run2pix.com/report/report_w.php?EventCode=20121216&Race=MA&sn=32"
    url_2013 = "https://www.run2pix.com/report/report_w.php?EventCode=20131215&Race=MA&sn=57"
    url_2014 = "https://www.run2pix.com/report/report_w.php?EventCode=20141221&Race=MA&sn=86"
    url_2015 = "https://www.run2pix.com/report/report_w.php?EventCode=20151220&Race=MA&sn=111"
    url_2016 = "https://www.run2pix.com/report/report_w.php?EventCode=20161218&Race=MA&sn=136"
    url_2017 = "https://www.run2pix.com/report/report_w.php?EventCode=20171217&Race=MA&sn=161"
    url_list = [url_2011, url_2012, url_2013, url_2014, url_2015, url_2016, url_2017]
    df_list = []

yaojenkuo

get_time_group <- function(x) {
  sub_2_half <- 2.5 * 3600
  sub_3_hour <- 3 * 3600
  sub_3_half <- 3.5 * 3600
  sub_4_hour <- 4 * 3600
  if (x < sub_2_half) {
    return("sub 2:30")
  } else if (x < sub_3_hour) {
    return("sub 3:00")
  } else if (x < sub_3_half) {

yaojenkuo

library(ggplot2)
library(dplyr)
library(magrittr)
library(ggtheme)
library(scale)
library(plotly)

runner_by_year_gender <- df %>%
  group_by(year, gender) %>%
  summarise(finishers = n())

yaojenkuo

# get_specdata()
get_specdata <- function(dest_file, ex_dir) {
  specdata_url <- "https://storage.googleapis.com/jhu_rprg/specdata.zip"
  download.file(specdata_url, destfile = dest_file)
  unzip(dest_file, exdir = ex_dir)
}
get_specdata("~/Downloads/specdata.zip", "~")

yaojenkuo

# pollutantmean()
pollutantmean <- function(directory, pollutant, id = 1:332) {
  csv_files <- list.files(directory) # 使用內建函數 list.files() 建立出 CSV 檔案路徑
  csv_file_paths <- paste0(directory, csv_files)
  csv_file_paths <- csv_file_paths[id] # 依照輸入的 id 參數選擇性讀入
  df_list <- list()
  pollutant_vector <- vector()
  for (i in 1:length(csv_file_paths)) {
    df_list[[i]] <- read.csv(csv_file_paths[i])
    pollutant_vector <- c(pollutant_vector, df_list[[i]][, pollutant]) # 將讀入測站的污染物資料合併起來

yaojenkuo

# complete()
complete <- function(directory, id = 1:332) {
  csv_files <- list.files(directory) # 使用內建函數 list.files() 建立出 CSV 檔案路徑
  csv_file_paths <- paste0(directory, csv_files)
  csv_file_paths <- csv_file_paths[id] # 依照輸入的 id 參數選擇性讀入
  df_list <- list()
  nobs <- vector()
  for (i in 1:length(csv_file_paths)) {
    df_list[[i]] <- read.csv(csv_file_paths[i])
    is_complete <- complete.cases(df_list[[i]]) # 直接引用 complete.cases() 函數

yaojenkuo

# corr()
corr <- function(directory, threshold = 0) {
  nobs <- complete(directory)[, "nobs"] # 使用上一題已經定義好的 complete() 函數
  if (threshold > max(nobs)) { # 如果使用者輸入的門檻值超過所有測站的最大完整觀測值
    return(NULL)
  } else {
    df_to_read <- nobs > threshold
    csv_files <- list.files(directory) # 使用內建函數 list.files() 建立出 CSV 檔案路徑
    csv_file_paths <- paste0(directory, csv_files)
    csv_file_paths <- csv_file_paths[df_to_read] # 利用邏輯值選出大於等於門檻的測站

yaojenkuo

# get_hospital_data()
get_hospital_data <- function(dest_file, ex_dir) {
  specdata_url <- "https://storage.googleapis.com/jhu_rprg/hospital_data.zip"
  download.file(specdata_url, destfile = dest_file)
  unzip(dest_file, exdir = ex_dir)
}
get_hospital_data("~/Downloads/hospital_data.zip", "~/hospital_data")

yaojenkuo

# To draw a histogram
library(ggplot2)

file_path <- "~/hospital_data/outcome-of-care-measures.csv"
outcome_of_care_measures <- read.csv(file_path, stringsAsFactors = FALSE)
outcome_of_care_measures[, 11] <- as.numeric(outcome_of_care_measures[, 11])
hist(outcome_of_care_measures[, 11], xlab = "30 Day Death Mortality Rates From Heart Attack", col = rgb(1, 0, 0, 0.5), main = "", breaks = 40) # base plotting system
ggplot(outcome_of_care_measures, aes(x = Hospital.30.Day.Death..Mortality..Rates.from.Heart.Attack)) +
  geom_histogram(bins = 40, fill = rgb(1, 0, 0, 0.5)) +
  xlab("30 Day Death Mortality Rates From Heart Attack") +

yaojenkuo

# best()
best <- function(state, outcome) {
  library(dplyr)
  library(magrittr)
  
  file_path <- "~/hospital_data/outcome-of-care-measures.csv"
  outcome_of_care_measures <- read.csv(file_path, stringsAsFactors = FALSE)
  outcome_col_idx <- c(11, 17, 23)
  for (i in outcome_col_idx) {
    outcome_of_care_measures[, i] <- suppressWarnings(as.numeric(outcome_of_care_measures[, i]))