計算 1 df 中的坐標向量與其他 df 中的單個坐標之間的距離

Question

假設我有以下兩個數據框（行不均勻）

set.seed(1999)
dfA <- data.frame(x = rpois(10,2), y = rpois(10,2), z = rpois(10,2), q = rpois(10,2), t = rpois(10,2))

set.seed(24)
dfB <- data.frame(a = rpois(10,2), b = rpois(10,2), c = rpois(10,2), d = rpois(10,2), e = rpois(10,2))

set.seed(10)
Dx <- sample.int(5)
set.seed(6)
Dy <- sample.int(5)

Dx <- as.data.frame(Dx)
Dx <- as.data.frame(transpose(Dx))
Dy <- as.data.frame(Dy)
Dy <- as.data.frame(transpose(Dy))

dfAB <- map2_df(dfA, dfB, str_c, sep=",") %>%
  rename_all(~ str_c('C', seq_along(.)))
dfXY <- map2_df(Dx, Dy, str_c, sep=",") %>%
  rename_all(~ str_c('C', seq_along(.)))

現在我有 2 個坐標數據集（dfAB 5 個變量，每個變量有 10 個觀察值，數據集 dfXY 5 個變量有 1 個觀察值）。

我想要的是一個新的數據框，其中包含 dfXY 變量 1 的觀察值與 dfAB 變量 1 中的每個單獨觀察值之間的距離、dfXY 變量 2 的觀察值 1 與 dfAB 變量 2 中的每個單獨觀察值之間的距離等.

如果我們有

dfAB                          dfXY
3,1   3,2  ...       3,5  1,2  2,1  5,4  4,3   
2,1   3,1                  
2,3   1,2               
...   ...            

那么新的數據幀應該包含以下距離：a) 3,5 & 3,1 b) 3,5 & 2,1 c) 3,5 & 2,3 等等...
以及之間的距離：a) 1,2 & 3,2 b) 1,2 & 3,1 c) 1,2 & 1,2 等等。
依此類推，所需的輸出格式為：

     dfABXY
C1:        C2:        C3:        C4:         C5 etc...
4.000000   2.000000   1.000000   3.605551
4.123106   2.236068   1.414214   3.605551              
2.236068   0.000000   2.236068   4.242641
3.162278   0.000000   1.000000   3.605551
3.162278   1.414214   2.000000   2.000000
etc...     etc...     etc...     etc...

但是，當我使用：

distances <- map2_df(
  dfAB,
  dfXY,
  ~ sqrt((.x$x - .y$x)^2 + (.x$y - .y$y)^2)
)

我收到錯誤Error in .x$x : $ operator is invalid for atomic vectors

我想我需要使用類似for(i in seq_along())函數但我不知道如何合並~ sqrt((.x$x - .y$x)^2 + (.x$y - .y$y)^2)

distance <- for(i in seq_along(dfXY)){
  dfAB[,i] <- dfAB[,i] [WHAT TO PUT HERE]

請注意，我的實際數據集有數千行和列，因此我手動調用行/列是不可行的。

任何幫助深表感謝

Answer 1

使用提供的示例數據，我來到了這個data.table -approach

library( data.table )
#set as data.tables
setDT(dfAB)
setDT(dfXY)
#melt to long format
dt1 <- melt( dfAB, measure.vars = names(dfAB) )[, c("x_AB","y_AB") := lapply( tstrsplit( value, ","), as.numeric ) ]
dt2 <- melt( dfXY, measure.vars = names(dfXY) )[, c("x_XY","y_XY") := lapply( tstrsplit( value, ","), as.numeric ) ]
#update join to get the coordinates to calculate distance with (join on Cx-value)
dt1[ dt2, `:=`( x_XY = i.x_XY, y_XY = i.y_XY ), on = .(variable) ]
#calculate eucledian distances
dt1[, distances := sqrt( (x_XY - x_AB )^2 + (y_XY - y_AB)^2 ) ]

輸出

#     variable value x_AB y_AB x_XY y_XY distances
#  1:       C1   3,1    3    1    3    5  4.000000
#  2:       C1   2,1    2    1    3    5  4.123106
#  3:       C1   2,3    2    3    3    5  2.236068
#  4:       C1   2,2    2    2    3    5  3.162278
#  5:       C1   4,2    4    2    3    5  3.162278
#  6:       C1   2,4    2    4    3    5  1.414214
#  7:       C1   4,1    4    1    3    5  4.123106
#  8:       C1   3,3    3    3    3    5  2.000000
#  9:       C1   2,3    2    3    3    5  2.236068
# 10:       C1   3,1    3    1    3    5  4.000000
# 11:       C2   3,2    3    2    1    2  2.000000
# 12:       C2   3,1    3    1    1    2  2.236068
# 13:       C2   1,2    1    2    1    2  0.000000
# 14:       C2   1,2    1    2    1    2  0.000000
# 15:       C2   0,1    0    1    1    2  1.414214
# 16:       C2   1,4    1    4    1    2  2.000000
# 17:       C2   0,1    0    1    1    2  1.414214
# 18:       C2   1,0    1    0    1    2  2.000000
# 19:       C2   4,2    4    2    1    2  3.000000
# 20:       C2   5,1    5    1    1    2  4.123106
# 21:       C3   2,0    2    0    2    1  1.000000
# 22:       C3   3,2    3    2    2    1  1.414214
# 23:       C3   3,3    3    3    2    1  2.236068
# 24:       C3   1,1    1    1    2    1  1.000000
# 25:       C3   0,1    0    1    2    1  2.000000
# 26:       C3   6,3    6    3    2    1  4.472136
# 27:       C3   2,0    2    0    2    1  1.000000
# 28:       C3   4,2    4    2    2    1  2.236068
# 29:       C3   1,2    1    2    2    1  1.414214
# 30:       C3   3,0    3    0    2    1  1.414214
# 31:       C4   3,1    3    1    5    4  3.605551
# 32:       C4   3,1    3    1    5    4  3.605551
# 33:       C4   2,1    2    1    5    4  4.242641
# 34:       C4   3,1    3    1    5    4  3.605551
# 35:       C4   3,4    3    4    5    4  2.000000
# 36:       C4   2,1    2    1    5    4  4.242641
# 37:       C4   4,3    4    3    5    4  1.414214
# 38:       C4   0,2    0    2    5    4  5.385165
# 39:       C4   2,3    2    3    5    4  3.162278
# 40:       C4   2,5    2    5    5    4  3.162278
# 41:       C5   2,2    2    2    4    3  2.236068
# 42:       C5   3,1    3    1    4    3  2.236068
# 43:       C5   2,1    2    1    4    3  2.828427
# 44:       C5   3,1    3    1    4    3  2.236068
# 45:       C5   1,0    1    0    4    3  4.242641
# 46:       C5   2,0    2    0    4    3  3.605551
# 47:       C5   1,1    1    1    4    3  3.605551
# 48:       C5   1,1    1    1    4    3  3.605551
# 49:       C5   4,1    4    1    4    3  2.000000
# 50:       C5   2,1    2    1    4    3  2.828427
#     variable value x_AB y_AB x_XY y_XY distances

更新

輸出格式更改為寬

#create id's to cast on
dt1[, id := rowidv( dt1, cols = "variable" ) ]
#cast to wide
dcast( dt1, id~variable, value.var = "distances" )

#    id       C1       C2       C3       C4       C5
# 1:  1 4.000000 2.000000 1.000000 3.605551 2.236068
# 2:  2 4.123106 2.236068 1.414214 3.605551 2.236068
# 3:  3 2.236068 0.000000 2.236068 4.242641 2.828427
# 4:  4 3.162278 0.000000 1.000000 3.605551 2.236068
# 5:  5 3.162278 1.414214 2.000000 2.000000 4.242641
# 6:  6 1.414214 2.000000 4.472136 4.242641 3.605551
# 7:  7 4.123106 1.414214 1.000000 1.414214 3.605551
# 8:  8 2.000000 2.000000 2.236068 5.385165 3.605551
# 9:  9 2.236068 3.000000 1.414214 3.162278 2.000000
#10: 10 4.000000 4.123106 1.414214 3.162278 2.828427

一次性的（並且還刪除了第一列）

dcast( dt1, rowidv( dt1, cols = "variable" )~variable, value.var = "distances" )[, -1]