CODEWITHSUNDEEP
rmatrixfilter
SteffenT
SteffenT

Reputation: 119

Delete rows/columns in matrix when values are only 0 across multiple matrices

I have multiple n*n matrices where each cell either contains values 0 or 1. My goal is to delete all IDs after a certain ID, where rows and columns have only 0's in all matrices of my data set.

Here is some data:

m1 <- matrix(c(1,0,0,1,0,0,0,0,0,0,0,0), 4, 4)
m2 <- matrix(c(1,0,0,1,0,0,0,0,0,0,0,0), 4, 4)

In this example, I would like to check for every ID after 2 if the corresponding row and column shows only 0's for both matrices. Here, this is true for ID 2&3: I then want to delete row & column 3 in all matrices, but keep row & column 2.

As you can imagine, my data set is a lot bigger and this can't be done by hand.

I tried:

m1b <- m1 [which(rowSums(m1[3:4,]) > 0 & colSums(m1[3:4,]) > 0), ]

But this omits the first 2 rows and only checks one matrix. Any help is appreciated.

Upvotes: 0

Views: 88

Answers (1)

Quentin
Quentin

Reputation: 360

  1. You have to put the column condition after the comma
  2. Your row condition is on m1[3:4,] but you apply it on m1. which(rowSums(m1[3:4,]) > 0) returns 2 but it is the 4th row of m1, you need to add 2 to the results of which and tell you want to keep the 2 first rows by concatenate the results with 1 and 2

For one matrix, you can do:

m1[c(1, 2, 2 + which(rowSums(m1[3:4,]) > 0)), which(colSums(m1[3:4,]) > 0)]

or

m1[c(rep(TRUE, 2), rowSums(m1[3:4,]) > 0), colSums(m1[3:4,]) > 0]

Now, to get the conditions for multiple matrices, you can use & between vectors of boolean

columnToKeep <- (colSums(m1[3:4,]) > 0) & (colSums(m2[3:4,]) > 0)
rowToKeep <- (c(rep(TRUE, 2), rowSums(m1[3:4,]) > 0)) & (c(rep(TRUE, 2), rowSums(m2[3:4,]) > 0))
m1[rowToKeep, columnToKeep]
m2[rowToKeep, columnToKeep]

Upvotes: 1

Related Questions