R-将dist函数应用于组

Question

I am trying to apply the dist() function row wise in R but the result I get is as if it isn't grouping at all, it is simply applying dist() to all of my dataframe.

df2 %>% dplyr::group_by(X1) %>% dist()

Where df2 is my dataframe and I am just applying to the head for now, for simplicity. Essentially, each group contains coordinates (A,B) and I am trying to get the distance between each point.

Here is my dataframe:

   X1  A              B
1   1  12             0.0
2   1  18             0.0
3   1  18             1.0
4   1  13             0.0
5   1  18             4.0
6   1  18             0.0
7   1  18             5.0
8   1  18             0.0
9   1  18             0.0
10  2  73            -2.0
11  2  73            -0.5
12  2  74            -0.5
13  2  73             0.0
14  2  71            -1.0
15  2  75             0.0

My desired output is the lower triangular matrix of each group, here is an example:

Answer 1

Here's an example of creating distance matrices of the iris data set by species

results = list()

for(spec in unique(iris$Species)){
  temp = iris[iris$Species==spec, 1:4]
  results[[length(results)+1]] = dist(temp)
}
names(results) = unique(iris$Species)

You'll have to figure out what to do with it afterwords.

Answer 2

We can user purrr::map :

library(purrr)

df %>% 
  split(.$X1) %>% 
  map(~{
    dist(.x)
  }) -> distList

distList
#> $`1`
#>          1        2        3        4        5        6        7        8
#> 2 6.000000                                                               
#> 3 6.082763 1.000000                                                      
#> 4 1.000000 5.000000 5.099020                                             
#> 5 7.211103 4.000000 3.000000 6.403124                                    
#> 6 6.000000 0.000000 1.000000 5.000000 4.000000                           
#> 7 7.810250 5.000000 4.000000 7.071068 1.000000 5.000000                  
#> 8 6.000000 0.000000 1.000000 5.000000 4.000000 0.000000 5.000000         
#> 9 6.000000 0.000000 1.000000 5.000000 4.000000 0.000000 5.000000 0.000000
#> 
#> $`2`
#>          10       11       12       13       14
#> 11 1.500000                                    
#> 12 1.802776 1.000000                           
#> 13 2.000000 0.500000 1.118034                  
#> 14 2.236068 2.061553 3.041381 2.236068         
#> 15 2.828427 2.061553 1.118034 2.000000 4.123106

Data:

df <- read.table(text = 'X1  A              B
1   1  12             0.0
2   1  18             0.0
3   1  18             1.0
4   1  13             0.0
5   1  18             4.0
6   1  18             0.0
7   1  18             5.0
8   1  18             0.0
9   1  18             0.0
10  2  73            -2.0
11  2  73            -0.5
12  2  74            -0.5
13  2  73             0.0
14  2  71            -1.0
15  2  75             0.0', h = T)

Answer 3

Here's my code and the solution

require(dplyr)
df2 <- structure(list(X1 = c(1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 2L, 
2L, 2L, 2L, 2L, 2L), A = c(12L, 18L, 18L, 13L, 18L, 18L, 18L, 
18L, 18L, 73L, 73L, 74L, 73L, 71L, 75L), B = c(0, 0, 1, 0, 4, 
0, 5, 0, 0, -2, -0.5, -0.5, 0, -1, 0)), .Names = c("X1", "A", 
"B"), class = "data.frame", row.names = c("1", "2", "3", "4", 
"5", "6", "7", "8", "9", "10", "11", "12", "13", "14", "15"))
mydf <- df2 %>% group_by(X1) %>% summarise(distmatrix=list(dist(cbind(A,B))))
mydf
# # A tibble: 2 × 2
# X1 distmatrix
# <int>     <list>
#   1     1 <S3: dist>
#   2     2 <S3: dist>
mydf$distmatrix
# [[1]]
# 1        2        3        4        5        6        7        8
# 2 6.000000                                                               
# 3 6.082763 1.000000                                                      
# 4 1.000000 5.000000 5.099020                                             
# 5 7.211103 4.000000 3.000000 6.403124                                    
# 6 6.000000 0.000000 1.000000 5.000000 4.000000                           
# 7 7.810250 5.000000 4.000000 7.071068 1.000000 5.000000                  
# 8 6.000000 0.000000 1.000000 5.000000 4.000000 0.000000 5.000000         
# 9 6.000000 0.000000 1.000000 5.000000 4.000000 0.000000 5.000000 0.000000
# 
# [[2]]
# 1        2        3        4        5
# 2 1.500000                                    
# 3 1.802776 1.000000                           
# 4 2.000000 0.500000 1.118034                  
# 5 2.236068 2.061553 3.041381 2.236068         
# 6 2.828427 2.061553 1.118034 2.000000 4.123106