Error Merging Two Columns

Question

I'm trying to combine two columns, Previous$Col1 and Previous$Col3, into one data.frame.

This is what I tried to do:

x<-data.frame(Previous$Col1)
y<-data.frame(Previous$Col3)
z<-merge(x,y)

And for some reason I got this error on the console:

Error: cannot allocate vector of size 24.0 Gb

What's going wrong and what should I do to fix this?

How could a data frame with two columns with 80000ish rows take up 24 GB of memory?

Thanks!

The error says that you ran out of memory. It is pretty uncommon to "merge" two vectors. What is your desired output? — lmo, Jul 20 '17 at 17:22
https://stackoverflow.com/questions/5171593/r-memory-management-cannot-allocate-vector-of-size-n-mb — AidanGawronski, Jul 20 '17 at 17:23
Possible duplicate of [R memory management / cannot allocate vector of size n Mb](https://stackoverflow.com/questions/5171593/r-memory-management-cannot-allocate-vector-of-size-n-mb) — AidanGawronski, Jul 20 '17 at 17:23
Is this one column you are trying to merge? Might work to append and then remove dupes.... — pyll, Jul 20 '17 at 17:24
But why did I run out of memory? I'm only trying to merge two 80000-row columns? How could that possibly be a 24 GB vector? — RaiderNAYSHUN, Jul 20 '17 at 17:24
@RaiderNAYSHUN - That's something `Rstudio` is not good at, if you are using it. Check this [comment](https://stackoverflow.com/questions/44894649/optimizing-apply-in-r#comment76839667_44895521). — Chetan Arvind Patil, Jul 20 '17 at 17:27
We can't know why you ran out of memory without seeing your data. It looks like `merge` on two vectors is equivalent to `expand.grid`, as seen here: ` merge(1:3, 2:4)` and `expand.grid(1:3, 2:4)`. So you would end up with a Cartesian product of the vectors, which is 6.4 billion rows with two columns. — lmo, Jul 20 '17 at 17:29
@Imo, my desired output is just a dataframe with the two columns combined — RaiderNAYSHUN, Jul 20 '17 at 17:29

Florian · Accepted Answer · 2017-07-20T17:34:33.397

You are creating a full cartesian product, which has 80000*80000 rows and two columns, that is a total of 1.28e+10 elements (about 51GB if I am correct). What are you trying to accomplish with your merge?

> x<-data.frame(a= c("a","b"))
> y<-data.frame(b= c(1,2,3))
> z<-merge(x,y)
> x
  a
1 a
2 b
> y
  b
1 1
2 2
3 3
> z
  a b
1 a 1
2 b 1
3 a 2
4 b 2
5 a 3
6 b 3

You could do data.frame(Col3 = Previous$Col3, Col1= Previous$Col1) to achieve what you want.

score 1 · Answer 2 · answered Jul 20 '17 at 17:36

Try using bind_cols from the dplyr package or cbind from base R.

bind_cols(Previous$Col1,Previous$Col3)

or

cbind(Previous$Col1,Previous$Col3)

Additionally, since these columns come from the same original data.frame. select() from the dplyr package could be used:

select(Previous,Col1,Col3)

Error Merging Two Columns

2 Answers2