Initializing data.frames()

Question

Is there a quick way to initialize an empty data frame? If you know what the dimensions will be? For example:

Suppose I would like a blank data frame that has 100 rows and 10:

x <- data.frame(1:100,2,3,4,5,6,7,8,9,10) 
dim(x) ## that's right

But suppose I want something like 300 columns? How do I quickly initialize columns in a data.frame?

x <- data.frame(1:100,2,3,4,5 ....) ## *cries*

@ZheyuanLi marking duplicates, 5 years later. Bravo – Brandon Bertelsen Nov 25 '16 at 18:24 — Brandon Bertelsen, Nov 25 '16 at 18:24

score 77 · Accepted Answer · answered Feb 01 '11 at 22:59

77

> df <- data.frame(matrix(ncol = 300, nrow = 100))
> dim(df)
[1] 100 300

answered Feb 01 '11 at 22:59

Gavin Simpson

How do I add column names for only some columns? – Herman Toothrot Nov 24 '15 at 11:55
1

`colnames(df) = paste0("a", 1:300)` – Mark Bower Aug 10 '18 at 18:47
@MarkBower +1 I would generally use `names()` for a data frame, however. – Gavin Simpson Aug 13 '18 at 16:15

score 36 · Answer 2 · answered Feb 01 '11 at 22:58

36

I always just convert a matrix:

x <- as.data.frame(matrix(nrow = 100, ncol = 10))

answered Feb 01 '11 at 22:58

Matt Parker

Thanks Matt! Looks like Gavin beat you to the type :P – Brandon Bertelsen Feb 01 '11 at 23:13
1

@Brandon actually @Matt beat me to it. I blame adding the `dim(df)` output for that ;-) – Gavin Simpson Feb 01 '11 at 23:16
1

Hmm, you proved the dimensions and actually used the number of columns specified in the question; what counts more toward bestfulness? It's probably unwise to contribute toward my ability to edit other people's posts, anyway... – Matt Parker Feb 01 '11 at 23:21

2 Answers2