我需要元素运算,而不是通常的矩阵乘法。以下工作正常:
# this works
Bmat <- structure(c(3L, 3L, 10L, 3L, 4L, 10L, 5L, 8L, 8L, 8L, 3L, 8L, 8L, 2L, 6L, 10L, 2L, 8L, 3L, 9L), .Dim = c(10L, 2L))
yvec <- c(2, 2, 2, 2, 2, 2, 2, 2, 2, 2)
Bmat * yvec
# [,1] [,2]
# [1,] 6 6
# [2,] 6 16
# [3,] 20 16
# [4,] 6 4
# [5,] 8 12
# [6,] 20 20
# [7,] 10 4
# [8,] 16 16
# [9,] 16 6
# [10,] 16 18
然而,以下失败:
# this fails
Amat <- structure(c(1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0), .Dim = c(10L, 2L))
xvec <- structure(c(1.83475682418716, 1.48154883122634, 1, 1, 1, 1, 1, 1, 1, 1), .Dim = c(10L, 1L))
Amat * xvec
#Fehler in Amat * xvec : nicht passende Arrays
为什么?它与Bmat
是一个整数矩阵有关吗?如何使第二个代码有效?
答案 0 :(得分:2)
class(xvec)
[1] "matrix"
dim(xvec)
[1] 10 1
class(Amat)
[1] "matrix"
dim(Amat)
[1] 10 2
两个矩阵之间的元素乘法只有在它们具有相同维度时才有可能。因此,解决方案是将xvec
转换为向量。试试
Amat * c(xvec)
#OR
Amat * as.vector(xvec)