스크립트를 함수로 변환하기

R 함수 작성 입문

Richie Cotton

Data Evangelist at DataCamp

기본 함수 템플릿

my_fun <- function(arg1, arg2) {
  # Do something
}

시그니처

          function(arg1, arg2)

본문

                               {
  # Do something
}

library(readr)
test_scores_geography_raw <- read_csv("test_scores_geography.csv")

library(dplyr)
library(lubridate)
test_scores_geography_clean <- test_scores_geography_raw %>% 
  select(person_id, first_name, last_name, test_date, score) %>%
  mutate(test_date = mdy(test_date)) %>%
  filter(!is.na(score))

library(readr)
test_scores_english_raw <- read_csv("test_scores_english.csv")

library(dplyr)
library(lubridate)
test_scores_english_clean <- test_scores_english_raw %>% 
  select(person_id, first_name, last_name, test_date, score) %>%
  mutate(test_date = mdy(test_date)) %>%
  filter(!is.na(score))

library(readr)
test_scores_art_raw <- read_csv("test_scores_art.csv")

library(dplyr)
library(lubridate)
test_scores_art_clean <- test_scores_art_raw %>% 
  select(person_id, first_name, last_name, test_date, score) %>%
  mutate(test_date = mdy(test_date)) %>%
  filter(is.na(score))

library(readr)
test_scores_spanish_raw <- read_csv("test_scores_spanish.csv")

library(dplyr)
library(lubridate)
test_scores_spanish_clean <- test_scores_spanish_raw %>% 
  select(person_id, first_name, last_name, test_date, score) %>%
  mutate(test_date = mdy(test_date)) %>%
  filter(!is.na(score))

템플릿 만들기

import_test_scores <- function() {






}

본문에 스크립트 붙여넣기

import_test_scores <- function() {
  test_scores_geography_raw <- read_csv("test_scores_geography.csv")

  test_scores_geography_clean <- test_scores_geography_raw %>% 
    select(person_id, first_name, last_name, test_date, score) %>%
    mutate(test_date = mdy(test_date)) %>%
    filter(!is.na(score))
}

인자를 선택하기

import_test_scores <- function(filename) {         # <- only 1 argument
  test_scores_geography_raw <- read_csv("test_scores_geography.csv")

  test_scores_geography_clean <- test_scores_geography_raw %>% 
    select(person_id, first_name, last_name, test_date, score) %>%
    mutate(test_date = mdy(test_date)) %>%
    filter(!is.na(score))
}

고정값을 인자로 대체하기

import_test_scores <- function(filename) {
  test_scores_geography_raw <- read_csv(filename)  # <- replace specific filename

  test_scores_geography_clean <- raw_data %>% 
    select(person_id, first_name, last_name, test_date, score) %>%
    mutate(test_date = mdy(test_date)) %>%
    filter(!is.na(score))
}

변수명 일반화하기

import_test_scores <- function(filename) {
  test_scores_raw <- read_csv(filename)   # <- variable names generalized

  test_scores_clean <- test_scores_raw %>%       # <- variable names generalized
    select(person_id, first_name, last_name, test_date, score) %>%
    mutate(test_date = mdy(test_date)) %>%
    filter(!is.na(score))
}

마지막 할당 제거하기

import_test_scores <- function(filename) {
  test_scores_raw <- read_csv(filename)

  test_scores_raw %>%  # <- remove assignment
    select(person_id, first_name, last_name, test_date, score) %>%
    mutate(test_date = mdy(test_date)) %>%
    filter(!is.na(score))
}

함수 사용하기

test_scores_geography <- import_test_scores("test_scores_geography.csv")
test_scores_english <- import_test_scores("test_scores_english.csv")
test_scores_art <- import_test_scores("test_scores_art.csv")
test_scores_spanish <- import_test_scores("test_scores_spanish.csv")

sample()의 인자

x: 샘플링할 값들의 벡터
size: x에서 샘플링할 횟수
replace: 복원 추출 여부
prob: x 각 값의 가중치 벡터(합계 1)

연습해 봅시다!

R 함수 작성 입문