Removing specific rows from a dataframe
I have a data frame e.g.:
sub day
1 1
1 2
1 3
1 4
2 1
2 2
2 3
2 4
3 1
3 2
3 3
3 4
and I would like to remove specific rows that can be identified by the combination of sub and day. For example say I wanted to remove rows where sub='1' and day='2' and sub=3 and day='4'. How could I do this? I realise that I could specify the row numbers, but this needs to be applied to开发者_开发百科 a huge dataframe which would be tedious to go through and ID each row.
DF[ ! ( ( DF$sub ==1 & DF$day==2) | ( DF$sub ==3 & DF$day==4) ) , ] # note the ! (negation)
Or if sub is a factor as suggested by your use of quotes:
DF[ ! paste(sub,day,sep="_") %in% c("1_2", "3_4"), ]
Could also use subset:
subset(DF, ! paste(sub,day,sep="_") %in% c("1_2", "3_4") )
(And I endorse the use of which
in Dirk's answer when using "[" even though some claim it is not needed.)
This boils down to two distinct steps:
- Figure out when your condition is true, and hence compute a vector of booleans, or, as I prefer, their indices by wrapping it into
which()
- Create an updated
data.frame
by excluding the indices from the previous step.
Here is an example:
R> set.seed(42)
R> DF <- data.frame(sub=rep(1:4, each=4), day=sample(1:4, 16, replace=TRUE))
R> DF
sub day
1 1 4
2 1 4
3 1 2
4 1 4
5 2 3
6 2 3
7 2 3
8 2 1
9 3 3
10 3 3
11 3 2
12 3 3
13 4 4
14 4 2
15 4 2
16 4 4
R> ind <- which(with( DF, sub==2 & day==3 ))
R> ind
[1] 5 6 7
R> DF <- DF[ -ind, ]
R> table(DF)
day
sub 1 2 3 4
1 0 1 0 3
2 1 0 0 0
3 0 1 3 0
4 0 2 0 2
R>
And we see that sub==2
has only one entry remaining with day==1
.
Edit The compound condition can be done with an 'or' as follows:
ind <- which(with( DF, (sub==1 & day==2) | (sub=3 & day=4) ))
and here is a new full example
R> set.seed(1)
R> DF <- data.frame(sub=rep(1:4, each=5), day=sample(1:4, 20, replace=TRUE))
R> table(DF)
day
sub 1 2 3 4
1 1 2 1 1
2 1 0 2 2
3 2 1 1 1
4 0 2 1 2
R> ind <- which(with( DF, (sub==1 & day==2) | (sub==3 & day==4) ))
R> ind
[1] 1 2 15
R> DF <- DF[-ind, ]
R> table(DF)
day
sub 1 2 3 4
1 1 0 1 1
2 1 0 2 2
3 2 1 1 0
4 0 2 1 2
R>
Here's a solution to your problem using dplyr's filter
function.
Although you can pass your data frame as the first argument to any dplyr function, I've used its %>%
operator, which pipes your data frame to one or more dplyr functions (just filter in this case).
Once you are somewhat familiar with dplyr, the cheat sheet is very handy.
> print(df <- data.frame(sub=rep(1:3, each=4), day=1:4))
sub day
1 1 1
2 1 2
3 1 3
4 1 4
5 2 1
6 2 2
7 2 3
8 2 4
9 3 1
10 3 2
11 3 3
12 3 4
> print(df <- df %>% filter(!((sub==1 & day==2) | (sub==3 & day==4))))
sub day
1 1 1
2 1 3
3 1 4
4 2 1
5 2 2
6 2 3
7 2 4
8 3 1
9 3 2
10 3 3
One simple solution:
cond1 <- df$sub == 1 & df$day == 2
cond2 <- df$sub == 3 & df$day == 4
df <- df[!(cond1 | cond2),]
精彩评论