tidyr

整齐的数据特性：

1. 每一列都是一个变量
2. 每一行都是一个观测值

tidyr 四大常用函数

1. gather() 使“宽”数据变成长数据
2. spread() 使“长”数据变成宽数据
3. separate() 将单个列拆分为多个列
4. unite() 将多个列组合成一个列

导入数据

library("lubridate")

ec2 <- ggplot2::economics %>% tbl_df() %>% 
  transmute(year = year(date), 
  month = month(date), rate = uempmed) %>%
  filter(year > 2005) %>% spread(year, rate)

image.png

1. gather()

gather 函数主要四个参数

1. data :数据集
2. key ：列明
3. value ：原来值的新的列名
4. ...: 需要聚集的变量，删除前面加-

gather(ec2, key=year, value=unemp, `2006`:`2015`)

image.png

# convert=TRUE, 把year 变量的字符串转成数值型，
# na.rm 删除缺失值
economics_2 <- gather(ec2, year,rate,`2006`:`2015`, convert = TRUE, na.rm = TRUE)

2. spread

• 是gather 函数的逆运算

weather <- dplyr::tibble(
  day = rep(1:3,2),
  obs = rep(c("temp","rain"),each = 3),
  val = c(c(23,22,20),c(0,0,5))
  )

image.png

spread(weather, key = obs, value = val)

image.png

3. separate

trt <- dplyr::tibble( var = paste0(rep(c("beg", "end"),
       each=3),"_",rep(c("a","b","c"))),
       val = c(1,4,2,10,5,11)
)

separate(trt,var,c("time","treatment"),"_")

image.png

4. unite

a <- separate(trt,var,c("time","treatment"),"_")
a %>% unite(var,time,treatment,sep="_")

image.png