linux awk命令实现对数据的每一列进行求和、求平均(Linux awk command can sum and average each column of data)

1、测试数据

root@PC1:/home/test# ls
test.txt
root@PC1:/home/test# cat test.txt
3 4 2 9
1 3 5 4
3 7 8 4
2 3 4 6

2、对每一列数据进行求和

root@PC1:/home/test# ls
test.txt
root@PC1:/home/test# cat test.txt
3 4 2 9
1 3 5 4
3 7 8 4
2 3 4 6
root@PC1:/home/test# for i in `head -n 1 test.txt | awk '{print NF}' | xargs seq`; do awk -v a=$i 'BEGIN{sum = 0} {sum += $a} END{print sum}' test.txt >> sum.txt; done
root@PC1:/home/test# ls
sum.txt  test.txt
root@PC1:/home/test# cat sum.txt  ##求和结果保存为列的形式
9
17
19
23
root@PC1:/home/test# paste -d " " -s sum.txt   ## 将列的形式转换为行的形式
9 17 19 23

3、对每一列求平均

root@PC1:/home/test# ls
test.txt
root@PC1:/home/test# cat test.txt
3 4 2 9
1 3 5 4
3 7 8 4
2 3 4 6
root@PC1:/home/test# for i in `head -n 1 test.txt | awk '{print NF}' | xargs seq `; do awk -v a=$i 'BEGIN{sum = 0} {sum += $a} END {print sum/NR}' test.txt >> mean.txt; done
root@PC1:/home/test# ls
mean.txt  test.txt
root@PC1:/home/test# cat mean.txt
2.25
4.25
4.75
5.75
root@PC1:/home/test# paste -d " " -s mean.txt   ## 每一列求平均结果
2.25 4.25 4.75 5.75

4、R语言实现

dir()
dat <- read.table("test.txt", header = F)
dat
apply(dat, 2, sum)
apply(dat, 2, mean)
> dir()
[1] "test.txt"
> dat <- read.table("test.txt", header = F)  ## 读取测试数据
> dat
  V1 V2 V3 V4
1  3  4  2  9
2  1  3  5  4
3  3  7  8  4
4  2  3  4  6
> apply(dat, 2, sum)     ## 求每一列的和
V1 V2 V3 V4 
 9 17 19 23 
> apply(dat, 2, mean)    ## 求每一列的平均数
  V1   V2   V3   V4 
2.25 4.25 4.75 5.75 
————————

1. Test data

root@PC1:/home/test# ls
test.txt
root@PC1:/home/test# cat test.txt
3 4 2 9
1 3 5 4
3 7 8 4
2 3 4 6

2. Sum each column of data

root@PC1:/home/test# ls
test.txt
root@PC1:/home/test# cat test.txt
3 4 2 9
1 3 5 4
3 7 8 4
2 3 4 6
root@PC1:/home/test# for i in `head -n 1 test.txt | awk '{print NF}' | xargs seq`; do awk -v a=$i 'BEGIN{sum = 0} {sum += $a} END{print sum}' test.txt >> sum.txt; done
root@PC1:/home/test# ls
sum.txt  test.txt
root@PC1:/home/test# cat sum.txt  ##求和结果保存为列的形式
9
17
19
23
root@PC1:/home/test# paste -d " " -s sum.txt   ## 将列的形式转换为行的形式
9 17 19 23

3. Average each column

root@PC1:/home/test# ls
test.txt
root@PC1:/home/test# cat test.txt
3 4 2 9
1 3 5 4
3 7 8 4
2 3 4 6
root@PC1:/home/test# for i in `head -n 1 test.txt | awk '{print NF}' | xargs seq `; do awk -v a=$i 'BEGIN{sum = 0} {sum += $a} END {print sum/NR}' test.txt >> mean.txt; done
root@PC1:/home/test# ls
mean.txt  test.txt
root@PC1:/home/test# cat mean.txt
2.25
4.25
4.75
5.75
root@PC1:/home/test# paste -d " " -s mean.txt   ## 每一列求平均结果
2.25 4.25 4.75 5.75

4. R language implementation

dir()
dat <- read.table("test.txt", header = F)
dat
apply(dat, 2, sum)
apply(dat, 2, mean)
> dir()
[1] "test.txt"
> dat <- read.table("test.txt", header = F)  ## 读取测试数据
> dat
  V1 V2 V3 V4
1  3  4  2  9
2  1  3  5  4
3  3  7  8  4
4  2  3  4  6
> apply(dat, 2, sum)     ## 求每一列的和
V1 V2 V3 V4 
 9 17 19 23 
> apply(dat, 2, mean)    ## 求每一列的平均数
  V1   V2   V3   V4 
2.25 4.25 4.75 5.75