補(bǔ)作業(yè)來la
dplyr五個(gè)基礎(chǔ)函數(shù)
- mutate()
- select()
- filter()
- arrange()
- summarise()
library(dplyr)
test <- iris[c(1:2,51:52,101:102),]
view(test)
mutate(test, new = Sepal.Length * Sepal.Width) #mutate(),新增列 ????♀?只新增是不保存的髓介，要再賦值一下
# select(),按列篩選
#按列號
select(test,1)
select(test,c(1,5))
select(test,Sepal.Length)
#按列名
select(test, Petal.Length, Petal.Width)
vars <- c("Petal.Length", "Petal.Width")
select(test,one_of(vars))
#??不知道這個(gè)one of是啥，我還搞笑的試了下two of ????
> select(test,two_of(vars))
錯誤: 沒有"two_of"這個(gè)函數(shù)
Run `rlang::last_error()` to see where the error occurred.
#
#選擇字符向量中的列嘹屯，select中不能直接使用字符向量篩選，需要使用one_of函數(shù)
vars <- c("Petal.Length", "Petal.Width")
select(iris, one_of(vars))
#返回指定字符向量之外的列 select(iris, -one_of(vars))
#返回所有列蹬昌，一般調(diào)整數(shù)據(jù)集中變量順序時(shí)使用 select(iris, everything())
#調(diào)整列順序安拟，把Species列放到最前面 select(iris, Species, everything())
#.filter()篩選行
#按給定的邏輯判斷篩選出符合要求的子數(shù)據(jù)集（表示AND時(shí)要使用&或者直接使用逗號）
filter(test, Species == "setosa")
filter(test, Species == "setosa"&Sepal.Length > 5 )
filter(test, Species %in% c("setosa","versicolor"))
# %in%：匹配，相當(dāng)于R中的match函數(shù)往枷，其表達(dá)的意思是左邊的元素在右邊的向量中是否存在登舞，如果存在則返回TRUE贰逾，否則返回FALSE
> filter(test, Species %in% c("setosa","versicolor"))
Sepal.Length Sepal.Width Petal.Length Petal.Width
1 5.1 3.5 1.4 0.2
2 4.9 3.0 1.4 0.2
3 7.0 3.2 4.7 1.4
4 6.4 3.2 4.5 1.5
Species new
1 setosa 17.85
2 setosa 14.70
3 versicolor 22.40
4 versicolor 20.48
#arrange(),按某1列或某幾列對整個(gè)表格進(jìn)行排序
arrange(test, Sepal.Length)#默認(rèn)從小到大排序
arrange(test, desc(Sepal.Length))#用desc從大到小
summarise()：匯總
summarise(test, mean(Sepal.Length), sd(Sepal.Length))# 計(jì)算Sepal.Length的平均
> summarise(test,mean(new),sd(new))
mean(new) sd(new)
1 18.64667 3.071597
group_by(test, Species)
summarise(group_by(test, Species),mean(Sepal.Length), sd(Sepal.Length))
> summarise(group_by(test, Species),mean(new), sd(new))
# A tibble: 3 x 3
Species `mean(new)` `sd(new)`
<fct> <dbl> <dbl>
1 setosa 16.3 2.23
2 versicolor 21.4 1.36
3 virginica 18.2 3.63
dplyr兩個(gè)實(shí)用技能
#1：管道操作 %>% (快捷鍵: cmd/ctr + shift + M)
> test %>%
+ group_by(Species) %>%
+ summarise(mean(new), sd(new))
# A tibble: 3 x 3
Species `mean(new)` `sd(new)`
<fct> <dbl> <dbl>
1 setosa 16.3 2.23
2 versicolor 21.4 1.36
3 virginica 18.2 3.63
#??和summarise的區(qū)別??
#2：count統(tǒng)計(jì)某列的unique值
count(test,Species)
dplyr處理關(guān)系數(shù)據(jù)
#將2個(gè)表進(jìn)行連接，注意：不要引入factor
options(stringsAsFactors = F)
test1 <- data.frame(x = c('b','e','f','x'),
z = c("A","B","C",'D'),
stringsAsFactors = F)
test1
test2 <- data.frame(x = c('a','b','c','d','e','f'),
y = c(1,2,3,4,5,6),
stringsAsFactors = F)
test2
#(1) 內(nèi)連inner_join,取交集
inner_join(test1, test2, by = "x")
#(2) 左連left_join
left_join(test1, test2, by = 'x')
#(3) 全連full_join
#(4) 半連接：返回能夠與y表匹配的x表所有記錄semi_join
semi_join(x = test1, y = test2, by = 'x')
#(5) 反連接：返回?zé)o法與y表匹配的x表的所記錄anti_join
anti_join(x = test1, y = test2, by = 'x')
#(6) 簡單合并
#在相當(dāng)于base包里的cbind()函數(shù)和rbind()函數(shù)
test1 <- data.frame(x = c(1,2,3,4), y = c(10,20,30,40))
test1
test2 <- data.frame(x = c(5,6), y = c(50,60))
test2
test3 <- data.frame(z = c(100,200,300,400))
test3
bind_rows(test1, test2) #bind_rows()函數(shù)需要兩個(gè)表格列數(shù)相同
bind_cols(test1, test3) #bind_cols()函數(shù)則需要兩個(gè)數(shù)據(jù)框有相同的行數(shù)
最后一個(gè)dplyr處理關(guān)系數(shù)據(jù)看的有點(diǎn)亂菠秒，應(yīng)該 要先對數(shù)據(jù)有初步了解疙剑、預(yù)想需要得到的數(shù)據(jù)結(jié)構(gòu)-再去選擇某種方法
