Create an empty data.frame without specifying column names

Question

I would like to have something like this:

#initialize a empty data frame without saying what the column names are    
dataFrame = data.frame()

for(1:20){

Create a row 

dataFrame = rbind(dataFrame, row)

}

The rows created in the for loop will have all the same column names. So is it possible to not initialize the column names?

Update:

Thank you all for the help.
But I wasn't clear with my answer, sorry.
I would like to have the names of the columns, stated in the for loop.

I now get the following message:
Error in match.names(clabs, nmi) : names do not match previous names
And what I can see is that the spaces in the column names get changed to dots. Is there a way so the names don't change or change beforehand.

It is hard to answer the question without knowing why you need to create a data frame without column names because it is possible that R has a more powerful way of doing what you want without a for () loop. Don't assume that R does things the same way as other programming languages. Tell us what you are trying to accomplish, don't tell use how you want to accomplish it. — dcarlson
– dcarlson, Commented Feb 1, 2018 at 22:28

Terru_theTerror · Accepted Answer · 2018-02-02 08:09:04Z

4

Inizialize it like this:

dataFrame<-NULL
db2<-data.frame(a=c(1,1,0,1,0,0)) 
rbind(dataFrame,db2)
  a
1 1
2 1
3 0
4 1
5 0
6 0

From your question-code (changed a little bit to make it running):

for(i in 1:5)
{
   row<-c(a=i,b=i,c=i)
   dataFrame = rbind(dataFrame, row)
}

The output

DataFrame
    a b c
row 1 1 1
row 2 2 2
row 3 3 3
row 4 4 4
row 5 5 5

edited Feb 2, 2018 at 8:09

answered Feb 1, 2018 at 16:41

Terru_theTerror

5,0152 gold badges22 silver badges39 bronze badges

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

4 Comments

Cing Over a year ago

I will add rows to the data frame not columns

Terru_theTerror Over a year ago

rbind add rows, not columns

Cing Over a year ago

But can you also have the row something like this: row <- c(a=i, b=i, c=i)

Terru_theTerror Over a year ago

Yes, you can. In this case you assign columns names inside the for loop. Answer changed according to this.

IceCreamToucan · Accepted Answer · 2018-02-01 16:45:16Z

3

for(i in 1:20){

# create a row 

if(i == 1) dataFrame <- t(row)
else dataFrame <- rbind(dataFrame, row)

}

Really though, it would be better to initialize a vector for each column, populate the vectors with the loop, then create a dataframe after the loop.

answered Feb 1, 2018 at 16:45

IceCreamToucan

28.8k2 gold badges27 silver badges48 bronze badges

Comments

d.b · Accepted Answer · 2018-02-01 16:51:12Z

3

If you know the type beforehand, you could do

df1 = data.frame(INTS = integer(0),
                 CHRS = character(),
                 NUMS = numeric(0))

for(i in 1:5){
    row = setNames(data.frame(i, letters[i], i + 0.5), names(df1))
    df1 = rbind(df1, row)
}

df1
#  INTS CHRS NUMS
#1    1    a  1.5
#2    2    b  2.5
#3    3    c  3.5
#4    4    d  4.5
#5    5    e  5.5

answered Feb 1, 2018 at 16:51

d.b

32.6k6 gold badges46 silver badges90 bronze badges

1 Comment

Cing Over a year ago

You still give names to the columns beforehand, I want to do that in the for loop.

Claudio Paladini · Accepted Answer · 2018-02-01 16:47:00Z

2

Either NULL or convert a martix to a dataframe at the end of the loop:

df = data.frame(NULL)

for(i in 1:20){
  row = i
  df = rbind(df,row)
}
class(df)
> [1] "data.frame"

or

df = NULL

for(i in 1:20){
  row = i
  df = rbind(df,row)
}
class(df)
df = data.frame(df)
class(df)

answered Feb 1, 2018 at 16:47

Claudio Paladini

1,0281 gold badge12 silver badges23 bronze badges

Collectives™ on Stack Overflow

Create an empty data.frame without specifying column names

4 Answers 4

4 Comments

Comments

1 Comment

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

4 Answers 4

4 Comments

Comments

1 Comment

Comments

Your Answer

Sign up or log in

Post as a guest

Related