Nếu đó là công việc quan trọng - hãy tự mình làm lấy

library(tidyverse)

## ── Attaching packages ────────────────────────────────────────────────────── tidyverse 1.2.1 ──

## ✔ ggplot2 3.1.1       ✔ purrr   0.3.2  
## ✔ tibble  2.1.1       ✔ dplyr   0.8.0.1
## ✔ tidyr   0.8.3       ✔ stringr 1.4.0  
## ✔ readr   1.3.1       ✔ forcats 0.4.0

## ── Conflicts ───────────────────────────────────────────────────────── tidyverse_conflicts() ──
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag()    masks stats::lag()

1 Giới thiệu

Không có package nào làm cho tôi đọc nhiều như purrr. Vì nó quá hay, quá thông minh

2 as_mapper()

Chuyển 1 object thành 1 mapper function

Là 1 cách viết khác của hàm

f_add <- function(x, y){x+y}

addition <- as_mapper(~ .x + .y)
addition

## <lambda>
## function (..., .x = ..1, .y = ..2, . = ..1) 
## .x + .y
## attr(,"class")
## [1] "rlang_lambda_function"

f_add(3,4)

## [1] 7

addition(3, 4)

## [1] 7

map2_int(1:3, 4:6, addition)

## [1] 5 7 9

map2_int(1:3, 4:6, f_add)

## [1] 5 7 9

as_mapper(f_add)

## function(x, y){x+y}
## <bytecode: 0x55cabf439268>

Đơn giản hơn khi chuyển đổi 1 object thành 1 hàm

get_rsquared <- compose(as_mapper("r.squared"), summary, lm)

get_rsquared(Sepal.Length ~ Sepal.Width, data = iris)

## [1] 0.01382265

l <- list(rnorm(10),
          rnorm(100), 
          rnorm(1000))
pmap_dbl(list(l, 0.1, TRUE), ~ mean(..1, ..2, ..3))

## [1] -0.37707235  0.05169313 -0.02798717

3 rerun()

Chạy 1 hàm với n lần

train <- rerun(20, sample_frac(airquality, size = 0.8))
validation <- map(train, ~ anti_join(airquality, .x))

## Joining, by = c("Ozone", "Solar.R", "Wind", "Temp", "Month", "Day")
## Joining, by = c("Ozone", "Solar.R", "Wind", "Temp", "Month", "Day")
## Joining, by = c("Ozone", "Solar.R", "Wind", "Temp", "Month", "Day")
## Joining, by = c("Ozone", "Solar.R", "Wind", "Temp", "Month", "Day")
## Joining, by = c("Ozone", "Solar.R", "Wind", "Temp", "Month", "Day")
## Joining, by = c("Ozone", "Solar.R", "Wind", "Temp", "Month", "Day")
## Joining, by = c("Ozone", "Solar.R", "Wind", "Temp", "Month", "Day")
## Joining, by = c("Ozone", "Solar.R", "Wind", "Temp", "Month", "Day")
## Joining, by = c("Ozone", "Solar.R", "Wind", "Temp", "Month", "Day")
## Joining, by = c("Ozone", "Solar.R", "Wind", "Temp", "Month", "Day")
## Joining, by = c("Ozone", "Solar.R", "Wind", "Temp", "Month", "Day")
## Joining, by = c("Ozone", "Solar.R", "Wind", "Temp", "Month", "Day")
## Joining, by = c("Ozone", "Solar.R", "Wind", "Temp", "Month", "Day")
## Joining, by = c("Ozone", "Solar.R", "Wind", "Temp", "Month", "Day")
## Joining, by = c("Ozone", "Solar.R", "Wind", "Temp", "Month", "Day")
## Joining, by = c("Ozone", "Solar.R", "Wind", "Temp", "Month", "Day")
## Joining, by = c("Ozone", "Solar.R", "Wind", "Temp", "Month", "Day")
## Joining, by = c("Ozone", "Solar.R", "Wind", "Temp", "Month", "Day")
## Joining, by = c("Ozone", "Solar.R", "Wind", "Temp", "Month", "Day")
## Joining, by = c("Ozone", "Solar.R", "Wind", "Temp", "Month", "Day")

Kiểm tra quy mô mẫu validation

map_int(validation, nrow) %>% every(~.x == 31)

## [1] TRUE

4 map_if()

Chỉ áp dụng khi thỏa mãn điều kiện của hàm if, giữ nguyên đầu vào nếu không thỏa mãn điều kiện if

Kiểm định xem 1 biến có phân phối chuẩn hay không

map_if(iris, is.numeric, shapiro.test)

## $Sepal.Length
## 
##  Shapiro-Wilk normality test
## 
## data:  .x[[i]]
## W = 0.97609, p-value = 0.01018
## 
## 
## $Sepal.Width
## 
##  Shapiro-Wilk normality test
## 
## data:  .x[[i]]
## W = 0.98492, p-value = 0.1012
## 
## 
## $Petal.Length
## 
##  Shapiro-Wilk normality test
## 
## data:  .x[[i]]
## W = 0.87627, p-value = 7.412e-10
## 
## 
## $Petal.Width
## 
##  Shapiro-Wilk normality test
## 
## data:  .x[[i]]
## W = 0.90183, p-value = 1.68e-08
## 
## 
## $Species
##   [1] setosa     setosa     setosa     setosa     setosa     setosa    
##   [7] setosa     setosa     setosa     setosa     setosa     setosa    
##  [13] setosa     setosa     setosa     setosa     setosa     setosa    
##  [19] setosa     setosa     setosa     setosa     setosa     setosa    
##  [25] setosa     setosa     setosa     setosa     setosa     setosa    
##  [31] setosa     setosa     setosa     setosa     setosa     setosa    
##  [37] setosa     setosa     setosa     setosa     setosa     setosa    
##  [43] setosa     setosa     setosa     setosa     setosa     setosa    
##  [49] setosa     setosa     versicolor versicolor versicolor versicolor
##  [55] versicolor versicolor versicolor versicolor versicolor versicolor
##  [61] versicolor versicolor versicolor versicolor versicolor versicolor
##  [67] versicolor versicolor versicolor versicolor versicolor versicolor
##  [73] versicolor versicolor versicolor versicolor versicolor versicolor
##  [79] versicolor versicolor versicolor versicolor versicolor versicolor
##  [85] versicolor versicolor versicolor versicolor versicolor versicolor
##  [91] versicolor versicolor versicolor versicolor versicolor versicolor
##  [97] versicolor versicolor versicolor versicolor virginica  virginica 
## [103] virginica  virginica  virginica  virginica  virginica  virginica 
## [109] virginica  virginica  virginica  virginica  virginica  virginica 
## [115] virginica  virginica  virginica  virginica  virginica  virginica 
## [121] virginica  virginica  virginica  virginica  virginica  virginica 
## [127] virginica  virginica  virginica  virginica  virginica  virginica 
## [133] virginica  virginica  virginica  virginica  virginica  virginica 
## [139] virginica  virginica  virginica  virginica  virginica  virginica 
## [145] virginica  virginica  virginica  virginica  virginica  virginica 
## Levels: setosa versicolor virginica

5 Phân tích tương quan

Tổ hợp 2 tên biến

library(tidystringdist) # Works since v0.1.2 
comb <- tidy_comb_all(names(airquality))
knitr::kable(comb)

V1	V2
Ozone	Solar.R
Ozone	Wind
Ozone	Temp
Ozone	Month
Ozone	Day
Solar.R	Wind
Solar.R	Temp
Solar.R	Month
Solar.R	Day
Wind	Temp
Wind	Month
Wind	Day
Temp	Month
Temp	Day
Month	Day

Bảng các hệ số tương quan và kiểm định ý nghĩa của các hệ số

bulk_cor <- pmap(comb, ~ cor.test(airquality[[.x]], airquality[[.y]])) %>% 
  map_df(broom::tidy) %>% 
  cbind(comb, .)

knitr::kable(bulk_cor, digits = 3)

V1	V2	estimate	statistic	p.value	parameter	conf.low	conf.high	method	alternative
Ozone	Solar.R	0.348	3.880	0.000	109	0.173	0.502	Pearson’s product-moment correlation	two.sided
Ozone	Wind	-0.602	-8.040	0.000	114	-0.706	-0.471	Pearson’s product-moment correlation	two.sided
Ozone	Temp	0.698	10.418	0.000	114	0.591	0.781	Pearson’s product-moment correlation	two.sided
Ozone	Month	0.165	1.781	0.078	114	-0.018	0.337	Pearson’s product-moment correlation	two.sided
Ozone	Day	-0.013	-0.141	0.888	114	-0.195	0.169	Pearson’s product-moment correlation	two.sided
Solar.R	Wind	-0.057	-0.683	0.496	144	-0.217	0.107	Pearson’s product-moment correlation	two.sided
Solar.R	Temp	0.276	3.444	0.001	144	0.119	0.419	Pearson’s product-moment correlation	two.sided
Solar.R	Month	-0.075	-0.906	0.366	144	-0.235	0.088	Pearson’s product-moment correlation	two.sided
Solar.R	Day	-0.150	-1.824	0.070	144	-0.305	0.012	Pearson’s product-moment correlation	two.sided
Wind	Temp	-0.458	-6.331	0.000	151	-0.575	-0.323	Pearson’s product-moment correlation	two.sided
Wind	Month	-0.178	-2.227	0.027	151	-0.328	-0.020	Pearson’s product-moment correlation	two.sided
Wind	Day	0.027	0.334	0.739	151	-0.132	0.185	Pearson’s product-moment correlation	two.sided
Temp	Month	0.421	5.703	0.000	151	0.281	0.543	Pearson’s product-moment correlation	two.sided
Temp	Day	-0.131	-1.619	0.108	151	-0.283	0.029	Pearson’s product-moment correlation	two.sided
Month	Day	-0.008	-0.098	0.922	151	-0.166	0.151	Pearson’s product-moment correlation	two.sided

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

Hướng dẫn sử dụng purrr package part 5

Nguyễn Ngọc Bình

30 Apr 2019

1 Giới thiệu

2 as_mapper()

3 rerun()

4 map_if()

5 Phân tích tương quan