CODEWITHSUNDEEP
rmatrixdataframe
pacomet
pacomet

Reputation: 5141

Change row order in a matrix/dataframe

I need to change/invert rows in my data frame, not transposing the data but moving the bottom row to the top and so on. If the data frame was:

1 2 3 
4 5 6
7 8 9

I need to convert to

7 8 9
4 5 6
1 2 3

I've read about sort() but I don't think it is what I need or I'm not able to find the way.

Upvotes: 27

Views: 37366

Answers (7)

Z Cao
Z Cao

Reputation: 51

Encounter this problem today and here I am providing another solution for your interests.

m <- matrix(1:9, ncol=3, byrow=TRUE)    
apply(m,2,rev)

Upvotes: 1

Rtist
Rtist

Reputation: 4205

I think this is the simplest way:

MyMatrix = matrix(1:20, ncol = 2)
MyMatrix[ nrow(MyMatrix):1, ]

If you want to reverse the columns, just do

MyMatrix[ , ncol(MyMatrix):1 ]

Upvotes: 6

zx8754
zx8754

Reputation: 56149

We can reverse the order of row.names (for data.frame only): 

# create data.frame
m <- matrix(1:9, ncol=3, byrow=TRUE)
df_m <- data.frame(m)

#reverse
df_m[rev(rownames(df_m)), ]

#   X1 X2 X3
# 3  7  8  9
# 2  4  5  6
# 1  1  2  3

Upvotes: 4

theogrost
theogrost

Reputation: 21

Veeery late, but this seems to be working fast, does not need any extra packages and is simple:

for(i in 1:ncol(matrix)) {matrix[,i] = rev(matrix[,i])}

I guess that for frequent use, one would make a function out of it. Tested with R v=3.3.1.

Upvotes: 2

Henrik
Henrik

Reputation: 3704

I would reverse the rows an index starting with the number of rows, along this line

revdata <-  thedata[dim(thedata)[1L]:1,]

Upvotes: 8

Holger Brandl
Holger Brandl

Reputation: 11192

You can reverse the order of a data.frame using the dplyr package:

iris %>% arrange(-row_number())

Or without using the pipe-operator by doing

arrange(iris, -row_number())

Upvotes: 28

Andrie
Andrie

Reputation: 179418

There probably are more elegant ways, but this works:

m <- matrix(1:9, ncol=3, byrow=TRUE)

# m[rev(seq_len(nrow(m))), ]  # Initial answer
m[nrow(m):1, ]
     [,1] [,2] [,3]
[1,]    7    8    9
[2,]    4    5    6
[3,]    1    2    3

This works because you are indexing the matrix with a reversed sequence of integers as the row index. nrow(m):1 results in 3 2 1.

Upvotes: 41

Related Questions