R语言数据处理——表格转换

1.定义表格：

df <- data.frame(x = c('A','B','C'),'2018' = c(1,3,4),'2019' = c(3,5,2),check.names = FALSE)
head(df)

控制台输出：

> df <- data.frame(x = c('A','B','C'),'2018' = c(1,3,4),'2019' = c(3,5,2),check.names = FALSE)
> head(df)
  x 2018 2019
1 A    1    3
2 B    3    5
3 C    4    2

注意一点的是，如果写成脚本形式，最好一步一步的点“Run”。

2.（方法1）将宽数据转换为长数据，将多行聚集成列

df_melt <- reshape2::melt(df,id.vars = "x",variable.name = "year",value.name = "value")
head(df_melt)

控制台输出：

> df_melt <- reshape2::melt(df,id.vars = "x",variable.name = "year",value.name = "value")
> head(df_melt)
  x year value
1 A 2018     1
2 B 2018     3
3 C 2018     4
4 A 2019     3
5 B 2019     5
6 C 2019     2

（方法2）

df_gather <- tidyr::gather(df,year,value,-x)
head(df_gather)

控制台输出：

> df_gather <- tidyr::gather(df,year,value,-x)
> head(df_gather)
  x year value
1 A 2018     1
2 B 2018     3
3 C 2018     4
4 A 2019     3
5 B 2019     5
6 C 2019     2

3.（方法1）将长数据转换为宽数据

df_cast <- reshape2::dcast(df_melt, x~year,value.var = 'value')
head(df_cast)

控制台输出：

> df_cast <- reshape2::dcast(df_melt, x~year,value.var = 'value')
> head(df_cast)
  x 2018 2019
1 A    1    3
2 B    3    5
3 C    4    2

（方法2）

df_spread <- tidyr::spread(df_gather,year,value)
head(df_spread)

 

4.变量的变换

方法1.

dat1 <- transform(df_melt,value2 = value*2)
head(dat1)

控制台输出：

> dat1 <- transform(df_melt,value2 = value*2)
> head(dat1)
  x year value value2
1 A 2018     1      2
2 B 2018     3      6
3 C 2018     4      8
4 A 2019     3      6
5 B 2019     5     10
6 C 2019     2      4

方法2.

dat2 <- transform(df_melt,value2 = ifelse(year=="2018",value*2,value+1))
head(dat2)

控制台输出：

> dat2 <- transform(df_melt,value2 = ifelse(year=="2018",value*2,value+1))
> head(dat2)
  x year value value2
1 A 2018     1      2
2 B 2018     3      6
3 C 2018     4      8
4 A 2019     3      4
5 B 2019     5      6
6 C 2019     2      3

方法3.

dat2 <- dplyr::mutate(df_melt,value2 = ifelse(year=="2018",value*2,value+1))
head(dat2)

5.表格的排序

升序：

dat_arrange1 <- dplyr::arrange(df_melt,value)
head(dat_arrange1)

控制台输出：

> dat_arrange1 <- dplyr::arrange(df_melt,value)
> head(dat_arrange1)
  x year value
1 A 2018     1
2 C 2019     2
3 B 2018     3
4 A 2019     3
5 C 2018     4
6 B 2019     5

降序：

dat_arrange2 <- dplyr::arrange(df_melt,desc(value))
head(dat_arrange2)

控制台输出：

> dat_arrange2 <- dplyr::arrange(df_melt,desc(value))
> head(dat_arrange2)
  x year value
1 B 2019     5
2 C 2018     4
3 B 2018     3
4 A 2019     3
5 C 2019     2
6 A 2018     1

全部的脚本：

# 表格的变换
df <- data.frame(x = c('A','B','C'),'2018' = c(1,3,4),'2019' = c(3,5,2),check.names = FALSE)
head(df)

df_melt <- reshape2::melt(df,id.vars = "x",variable.name = "year",value.name = "value")
head(df_melt)

df_gather <- tidyr::gather(df,year,value,-x)
head(df_gather)
  
df_cast <- reshape2::dcast(df_melt, x~year,value.var = 'value')
head(df_cast)

df_spread <- tidyr::spread(df_gather,year,value)
head(df_spread)
# 变量的变换
dat1 <- transform(df_melt,value2 = value*2)
head(dat1)

dat2 <- transform(df_melt,value2 = ifelse(year=="2018",value*2,value+1))
head(dat2)

dat2 <- dplyr::mutate(df_melt,value2 = ifelse(year=="2018",value*2,value+1))
head(dat2)

# 表格的排序
dat_arrange1 <- dplyr::arrange(df_melt,value)
head(dat_arrange1)

dat_arrange2 <- dplyr::arrange(df_melt,desc(value))
head(dat_arrange2)

操作：