[Johns Hopkins] R Programming 作業 Week 2 - Air Pollution

pollutantmean <- function(directory, pollutant, id = 1:332) {
        ## ‘directory‘ is a character vector of length 1 indicating
        ## the location of the CSV files

        ## ‘pollutant‘ is a character vector of length 1 indicating
        ## the name of the pollutant for which we will calculate the
        ## mean; either "sulfate" or "nitrate".

        ## ‘id‘ is an integer vector indicating the monitor ID numbers
        ## to be used

        ## Return the mean of the pollutant across all monitors list
        ## in the ‘id‘ vector (ignoring NA values)
        ## NOTE: Do not round the result!
}

complete <- function(directory, id = 1:332) {
        ## ‘directory‘ is a character vector of length 1 indicating
        ## the location of the CSV files

        ## ‘id‘ is an integer vector indicating the monitor ID numbers
        ## to be used
        
        ## Return a data frame of the form:
        ## id nobs
        ## 1  117
        ## 2  1041
        ## ...
        ## where ‘id‘ is the monitor ID number and ‘nobs‘ is the
        ## number of complete cases
}

corr <- function(directory, threshold = 0) {
        ## ‘directory‘ is a character vector of length 1 indicating
        ## the location of the CSV files

        ## ‘threshold‘ is a numeric vector of length 1 indicating the
        ## number of completely observed observations (on all
        ## variables) required to compute the correlation between
        ## nitrate and sulfate; the default is 0

        ## Return a numeric vector of correlations
        ## NOTE: Do not round the result!
}

# 設立get_specdata()
get_specdata <- function(dest_file) {  
  specdata_url <- "https://storage.googleapis.com/jhu_rprg/specdata.zip"   #擷取檔案下載的url
  download.file(specdata_url, destfile = dest_file)  　　　　　　　　　　    　#以download.file下載，destfile = 指定位置 *註:此處~會為R主程式的wd
  unzip(dest_file)　　　　　　　　　　　　　　　　　　　　                        #unzip檔案至Rstudio的wd
}
get_specdata("~/specdata.zip")

#可指定解壓位置的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)　　　　　　　　　　　　　　　　　　　　        #exdir為指定位置*註:此處~會為R主程式的wd
}
get_specdata("~/specdata.zip", "D:/R/Project")

pollutantmean <- function(directory,pollutant,id = 1:332) {
  CSV_files_dir <- list.files(directory, full.names = T)  #將茲目標料夾中的files，匯成list
  dataf <-data.frame() 
  for(i in id){
    dataf <- rbind(dataf,read.csv(CSV_files_dir[i]))      #rbind將for迴圈的資料綁成新row
  }
  mean(dataf[,pollutant],na.rm = T)                       #所有row的 指定column做計算
}

pollutantmean <- function(directory, pollutant, id= 1:332){
  pollutants = c()                                            #設立空vector用於接數據
  filenames = list.files(directory)                           #此處沒有用 full_name參數，只會有files name 

  for(i in id){
    filepath=paste(directory,"/" ,filenames[i], sep="")       #將檔名與路徑貼起來，製成完整路徑fliepath    
    data = read.csv(filepath, header = TRUE)                  #讀取目標檔案及其header，存至data   
    pollutants = c(pollutants, data[,pollutant])              #將每筆數據加長至vector中,存至pollutants
  }
  pollutants_mean = mean(pollutants, na.rm=TRUE)              #計算並存至pollutants_mean
  
  pollutants_mean                                             #回報
}

練習

pollutantmean("specdata", "sulfate", 1:10)
[1] 4.064

pollutantmean("specdata", "nitrate", 70:72)
[1] 1.706

pollutantmean("specdata", "sulfate", 34)
[1] 1.477

pollutantmean("specdata", "nitrate")
[1] 1.703

complete <- function(directory, id = 1:332) {
  CSV_files <- list.files(directory, full.names = TRUE)
  datadf <- data.frame()
  for (i in id) {
    moni_i <- read.csv(CSV_files[i])
    nobs <- sum(complete.cases(moni_i))      #complete.cases()可得是否為complete的邏輯vector，sum()加總True值
    tmpdf <- data.frame(i, nobs)             #將測站ID及其結果存成 df
    datadf <- rbind(datadf, tmpdf)           #將新的資料綁至新row
  }
  colnames(datadf) <- c("id", "nobs")        #將column賦名
  datadf                                     #回報
}

練習

查看指定感測器中，具有完整資訊的筆數
cc <- complete("specdata", c(6, 10, 20, 34, 100, 200, 310))   #cc5中有"id" "nobs" 兩columns
print(cc$nobs)                                                #nobs的 vector

[1] 228 148 124 165 104 460 232

查看指定感測器中，具有完整資訊的筆數
cc <- complete("specdata", 54)                               #cc中有"id" "nobs" 兩columns
print(cc$nobs)                                               #nobs的 vector
[1] 219

隨機抽樣查看10組感測器，具有完整資訊的筆數
set.seed(42)
cc <- complete("specdata", 332:1)                            #cc中有 "id" "nobs"兩columns *row是反讀，但此處沒差
use <- sample(332, 10)                                       #332中亂數取10個成  use vector
print(cc[use, "nobs"])                                       #第 use row 的 "nobs"

[1] 711 135  74 445 178  73  49   0 687 237

corr <- function(directory, threshold = 0) {                           #門檻defalut = 0
  CSV_files <- list.files(directory, full.names = TRUE)
  dat <- vector(mode = "numeric", length = 0)                          #設置空的numeric vector
  for (i in 1:length(CSV_files)) {
    moni_i <- read.csv(CSV_files[i])                                   #此處沒有指定id，直接以length讀長度
    csum <- sum((!is.na(moni_i$sulfate)) & (!is.na(moni_i$nitrate)))   #獲得兩側相都沒na測值的True數量
    if (csum > threshold) {                                            #超出門檻的
      tmp <- moni_i[which(!is.na(moni_i$sulfate)), ]                   #留下sulfate是True的
      submoni_i <- tmp[which(!is.na(tmp$nitrate)), ]                   #再留下nitrate是True的
      dat <- c(dat, cor(submoni_i$sulfate, submoni_i$nitrate))         #將cor()值綁長至dat vector 中
    }
  }
  dat
}

練習

從排序完成的相關係數中，隨機抽樣5組，並四捨五入至小數點下第四位
cr <- corr("specdata") 
cr <- sort(cr) 
set.seed(868) 
out <- round(cr[sample(length(cr), 5)], 4)
print(out)

[1] 0.2688 0.1127 -0.0085 0.4586 0.0447

資料完整數大於129筆的資料組數，其相關係數排序完成後隨機抽樣5組，並四捨五入至小數點下第四位
cr <- corr("specdata", 129) 
cr <- sort(cr) 
n <- length(cr) 
set.seed(197) 
out <- c(n, round(cr[sample(n, 5)], 4))
print(out)

[1] 243.0000   0.2540   0.0504  -0.1462  -0.1680   0.5969

資料完整度大於2000筆的資料組數，與資料完整度大於1000筆的資料，其相關係數排序完成後以四捨五入呈現至小數點下第四位
cr <- corr("specdata", 2000) 
n <- length(cr) 
cr <- corr("specdata", 1000) 
cr <- sort(cr)
print(c(n, round(cr, 4)))

[1]  0.0000 -0.0190  0.0419  0.1901