Remove row by row number in a dataframe

Question

I would like to create a very simple function that will take the dataframe and row number as arguemnts, and then return a dataframe without that row. So if I have:

1 ben
2 adrian
3 bill
4 matthew

After calling remove(3,df) I will get:

1 ben
2 adrian
3 matthew

Anyone can help out?

Are you aware of the fact that you can use e.g. df[-3,] to remove row 3 from the output? (or in case of multiple rows to be removed: df[-c(1:2),])? — talat
– talat, Commented Aug 15, 2014 at 11:08
@beginneR, he will need , drop=F too in order to keep it in a data.frame structure, but that's basically it — David Arenburg
– David Arenburg, Commented Aug 15, 2014 at 11:10
I want to permantently remove it from the dataframe and then save it to rda file — LucasSeveryn
– LucasSeveryn, Commented Aug 15, 2014 at 11:11
@LucasSeveryn, to permanently remove it, use df <- df[-3,, drop=FALSE] and then save it as you wish. — talat
– talat, Commented Aug 15, 2014 at 11:12

Dan · Accepted Answer · 2014-08-15 11:17:30Z

4

This function can be written as a one-liner (albeit without error checking):

remrow <- function(x, rows) x[-rows,, drop = FALSE]

Here's your data:

dat <- data.frame(A = 1:4, B = 5:8)
rownames(dat) <- c("ben", "adrian", "bill", "matthew")

remrow(dat, 2) will strip out the "adrian" row, while remrow(dat, 1:2) will remove both the "ben" and "adrian" rows.

Note that is.data.frame(remrow(dat, 1)) evaluates to TRUE.

edited Aug 15, 2014 at 11:17

answered Aug 15, 2014 at 11:12

Dan

5,4491 gold badge29 silver badges39 bronze badges

Sign up to request clarification or add additional context in comments.

Comments

Rich Scriven · Accepted Answer · 2014-08-15 15:53:21Z

3

It seems like this is all you really need

> removeRows <- function(rowNum, data) {
      newData <- data[-rowNum, , drop = FALSE]
      rownames(newData) <- NULL
      newData
  }
> dat
#        V2
# 1     ben
# 2  adrian
# 3    bill
# 4 matthew
> removeRows(3, dat)
#        V2
# 1     ben
# 2  adrian
# 3 matthew

answered Aug 15, 2014 at 15:53

Rich Scriven

99.8k11 gold badges190 silver badges252 bronze badges

Comments

akrun · Accepted Answer · 2014-08-15 14:44:52Z

1

If df is the dataset

fun1 <- function(data, rowN){
 indx <- !(seq_len(nrow(data)) %in% rowN)
 data1 <- data[indx,, drop=F]
 row.names(data1) <- seq_len(nrow(data1))
 data1
 }

fun1(df, 3)
#       V1
#1     ben
#2  adrian
#3 matthew

fun1(df,1:2)
#       V1
#1    bill
#2 matthew

edited Aug 15, 2014 at 14:44

answered Aug 15, 2014 at 10:56

akrun

891k38 gold badges590 silver badges700 bronze badges

4 Comments

David Arenburg Over a year ago

Why are you using row.names here? All he wants is a row number. It is quite possible that the row.names won't fit the row number. I don't think anything else but data[-rowN,, drop=F] is needed in this function

akrun Over a year ago

@David Arenburg, I changed row.names to 1:nrow(data). Yes, you are right. there may be cases with different rownames.

akrun Over a year ago

@beginneR. You are absolutely correct. Thanks for your comments. Updated the code.

Rich Scriven Over a year ago

rownames(data) <- NULL will reset the row names to a sequence also.

Collectives™ on Stack Overflow

Remove row by row number in a dataframe

3 Answers 3

Comments

Comments

4 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

3 Answers 3

Comments

Comments

4 Comments

Your Answer

Sign up or log in

Post as a guest

Related