本節(jié)來介紹dplyr中的重要函數(shù)count，summarise，group_by
count
count 統(tǒng)計觀察次數(shù)
library(tidyverse)
msleep %>%
count(order, sort = TRUE)
order n
<chr> <int>
1 Rodentia 22
2 Carnivora 12
3 Primates 12
4 Artiodactyla 6
5 Soricomorpha 5
也可以在一個count()語句中添加多個變量
msleep %>%
count(order, vore, sort = TRUE)
order vore n
<chr> <chr> <int>
1 Rodentia herbi 16
2 Carnivora carni 12
3 Primates omni 10
4 Artiodactyla herbi 5
summarize
dplyr 中的summarize函數(shù)使用直觀易讀的代碼對統(tǒng)計數(shù)據(jù)進(jìn)行匯總
msleep %>%
summarise(n = n(), average = mean(sleep_total), maximum = max(sleep_total))
## # A tibble: 1 x 3
## n average maximum
## <int> <dbl> <dbl>
## 1 83 10.4 19.9
group_by( )按分組進(jìn)行匯總
msleep %>%
group_by(vore) %>%
summarise(n = n(), average = mean(sleep_total), maximum = max(sleep_total))
## # A tibble: 5 x 4
## vore n average maximum
## <chr> <int> <dbl> <dbl>
## 1 carni 19 10.4 19.4
## 2 herbi 32 9.51 16.6
## 3 insecti 5 14.9 19.9
## 4 omni 20 10.9 18.0
## 5 <NA> 7 10.2 13.7
summarise( )幾乎適用于任何聚合函數(shù)，并允許進(jìn)行額外的算術(shù)運算：
- n() - 給出觀察次數(shù)
- n_distinct(var) - 給出唯一值的數(shù)量 var
- sum(var), max(var), min(var), ...
- mean(var), median(var), sd(var), IQR(var)
將平均 sleep_total 并除以 24，以獲得一天的睡眠量
msleep %>%
group_by(vore) %>%
summarise(avg_sleep_day = mean(sleep_total)/24)
## # A tibble: 5 x 2
## vore avg_sleep_day
## <chr> <dbl>
## 1 carni 0.432
## 2 herbi 0.396
## 3 insecti 0.622
## 4 omni 0.455
## 5 <NA> 0.424
summarise_all()需要一個函數(shù)作為參數(shù),它將應(yīng)用于所有列；示例代碼計算每列的平均值
msleep %>%
group_by(vore) %>%
summarise_all(mean, na.rm=TRUE)
## # A tibble: 5 x 11
## vore name genus order conservation sleep_total sleep_rem sleep_cycle
## <chr> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 carni NA NA NA NA 10.4 2.29 0.373
## 2 herbi NA NA NA NA 9.51 1.37 0.418
## 3 insecti NA NA NA NA 14.9 3.52 0.161
給每列的值加5
msleep %>%
group_by(vore) %>%
summarise_all(~mean(., na.rm = TRUE) + 5)
## vore name genus order conservation sleep_total sleep_rem sleep_cycle
## <chr> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 carni NA NA NA NA 15.4 7.29 5.37
## 2 herbi NA NA NA NA 14.5 6.37 5.42
## 3 insecti NA NA NA NA 19.9 8.52 5.16
summarise_if()
計算所有數(shù)字列的平均值
msleep %>%
group_by(vore) %>%
summarise_if(is.numeric, mean, na.rm=TRUE)
rename_if( )對列進(jìn)行重命名
msleep %>%
group_by(vore) %>%
summarise_if(is.numeric, mean, na.rm=TRUE) %>%
rename_if(is.numeric, ~paste0("avg_", .))
## vore avg_sleep_total avg_sleep_rem avg_sleep_cycle avg_awake
## <chr> <dbl> <dbl> <dbl> <dbl>
## 1 carni 10.4 2.29 0.373 13.6
## 2 herbi 9.51 1.37 0.418 14.5
## 3 insecti 14.9 3.52 0.161 9.06
summarise_at()
下面的代碼將返回平均含有單詞“睡眠”的所有列，并且還它們重命名為“AVG_ VAR"
msleep %>%
group_by(vore) %>%
summarise_at(vars(contains("sleep")), mean, na.rm=TRUE) %>%
rename_at(vars(contains("sleep")), ~paste0("avg_", .))
top_n( )
保留值最高的5個
msleep %>%
group_by(order) %>%
summarise(average = mean(sleep_total)) %>%
top_n(5)
保留值最低的5個
msleep %>%
group_by(order) %>%
summarise(average = mean(sleep_total)) %>%
top_n(-5)
示例代碼將保留average_sleep 的5 個最高值
msleep %>%
group_by(order) %>%
summarise(average_sleep = mean(sleep_total), max_sleep = max(sleep_total)) %>%
top_n(5, average_sleep)
## order average_sleep max_sleep
## <chr> <dbl> <dbl>
## 1 Afrosoricida 15.6 15.6
## 2 Chiroptera 19.8 19.9
## 3 Cingulata 17.8 18.1
## 4 Didelphimorphia 18.7 19.4
sample_frac()允許隨機選擇一部分行（此處為 10%）
msleep %>% sample_frac(.1)
喜歡的小伙伴歡迎關(guān)注我的公眾號
R語言數(shù)據(jù)分析指南，持續(xù)分享數(shù)據(jù)可視化的經(jīng)典案例及一些生信知識，希望對大家有所幫助
