R:如果前一行的最后一列元素为NA，则使当前行为NA

Question

我正在写一个函数，它将取向量V.Size中最大的元素并输出到N × N+1矩阵中。我的问题是当V.Size小于N*(N+1)。当这种情况发生时，矩阵从向量的顶部开始，而我希望它输出NA s。

例如:

# vector V.size is
V.size <- c(1,2,3,4,5,6)
# and N is
N <- 2
# then, the output matrix should be
   c1 c2 c3  
r1  6  5  4  
r2  3  2  1

当N*(N+1) > V.Size时，我希望V.Size填充它，直到V.Size用完，然后返回NA s而不是重新开始。

我试图解决这个问题是通过搜索一个元素何时比前一个元素大，并将其替换为NA。我尝试的解决方案返回错误:

Error in if (is.na(m)[(i - 1), (y + 1)]) { : argument is of length zero

下面是我的代码:

# Function Name: one
# Input: V.Size (a vector) and N
# Output: Matrix size N by N+1
# Code:
one <- function(x,y) {
  # sort the data, largest to smallest with N.A. last
  temp <- sort(x, decreasing = TRUE, na.last = TRUE)
  #creating the matrix
  m <- matrix(head(temp, (y*(y+1))), # only takes elements that fit in the matrix
    nrow = y,                     # number of rows = N
    ncol = (y+1),                 # number of columns = N+1
    byrow = TRUE)                 # filling it by row as instructed
  if (length(x) < (y*(y+1))) {    # if V.Size is smaller than the outputted matrix
    for (i in seq_len(y)) {       # for loop for columns 
      for (j in seq_len(y+1)) {   # for loop for rows
        if (m[i, j] > m[i,1]) {   # if the element is larger than the first in the row 
          m[i, j] = NA            # return NA
        }
# HERE IS WHERE THINGS FAIL:
        if (is.na(m)[(i-1), (y+1)]) { # if the last element in the previous row is NA 
         m[i, ] = NA                  # make current row NA
        }
      }
    }
  }
  # print the output
  m
}
# creating dummy data
V.Size <- c(1:10)
# choosing a dummy N
N = 5
one(V.Size, N)

我得到错误:Error in if (is.na(m)[(i - 1), (y + 1)]) { : argument is of length zero

Answer 1

这个怎么样?

V.size <- 1:6
N <- 3
matrix(sort(V.size, decreasing=TRUE)[1:(N*(N+1))], nrow=N, byrow=TRUE)
     [,1] [,2] [,3] [,4]
[1,]    6    5    4    3
[2,]    2    1   NA   NA
[3,]   NA   NA   NA   NA

Answer 2

我认为问题是在循环的第一次运行。当循环第一次运行时，不应该检查条件。即，当i-1 = -1或i=1时，则不检查。不会有以前的成员第一次运行!!