Dataset attribute headings
I am a beginner and I am trying something like this:
for (i in newTrain) {
count = 0
count = length(which(is.na(newTrain$i)))
names(-which(count>100))
}
but this isn't working at all for me.
Asked
Active
Viewed 127 times
-1
Jaap
- 81,064
- 34
- 182
- 193
Muhammad Wasif
- 145
- 1
- 2
- 10
-
1Welcome to StackOverflow! Please read the info about [how to ask a good question](http://stackoverflow.com/help/how-to-ask) and how to give a [reproducible example](http://stackoverflow.com/questions/5963269/how-to-make-a-great-r-reproducible-example/5963610). This will make it much easier for others to help you. – Jaap Oct 27 '16 at 11:47
-
newTrain[apply(newTrain, 2, function(x) sum(is.na(x))<=100)] – Lourdes Hernández Oct 27 '16 at 11:51
-
@LourdesHernández no need to use `apply` to `sum` on each `column`, there is a `colSums` function (see the posted answer) – Cath Oct 27 '16 at 11:52
1 Answers
3
We could first apply is.na for the entire dataframe and then sum the value of NAs for every column. Then select columns which have NA value less than 100.
newTrain[colSums(is.na(newTrain)) < 100]
Ronak Shah
- 377,200
- 20
- 156
- 213
-
1Its working...but it was missing a comma newTrain[,colSums(is.na(newTrain)) < 100] Thank you – Muhammad Wasif Oct 27 '16 at 12:01
-
@MuhammadWasif that is strange. It works for me with/without comma. – Ronak Shah Oct 27 '16 at 12:11
-
1@RonakShah It works on a `data.frame` but not on a `matrix`. In either case, you’ll need a `, drop = FALSE` at the end — otherwise this won’t work when only a single column remains. – Konrad Rudolph Oct 27 '16 at 12:25
-
@KonradRudolph I see, so `newTrain[,colSums(is.na(newTrain)) < 100, drop = FALSE]` would be perfect. Thanks. – Ronak Shah Oct 27 '16 at 12:38
-
Mr. Zheyuan Li .. please dont do this.. I was struck in this small problem since yesterday and he helped me and I find it great support – Muhammad Wasif Oct 27 '16 at 12:40
-
1