How to create an empty dataframe with specified columns AND number of rows in R?

Question

I have two dataframes that I want to be able to rbind together. They have the similar information in them, but not in the same order and not with the same column names (and this is a mix of strings, integers, and real numbers, so matrices will not work).

What I then need to do is convert one of the dataframes (we'll call it new_df) into the same structure as the other dataframe (we'll call it old_df).

I want to create an empty dataframe with the column structure of old_df AND the same number of rows as new_df.

I know I can create the first part of that with empty_df <- old_df[0,], but how can I specify the number of rows?

I know the number of rows I want to end up with, so I'd like to specify that. I cannot find this anywhere.

What I want is something like this (if this worked):

empty_df <- old_df[rep(0,nrow(new_df)),]

I tried:

empty_df <- old_df[rep(0,nrow(new_df)),]

This just does the same as old_df[0,] with 0 rows

empty_df <- old_df[0,]

empty_df$ID <- new_df$ids

Obviously that doesn't work as I am trying to add a different number of rows

something like this? as.data.frame(matrix(,nrow(df),ncol(df))) — Wimpel
– Wimpel, Commented May 10, 2019 at 18:47
This is an XY Problem where you ask help on your y solution of empty data frame but not the x problem of why you need such an object. My guess is you will then iterate through loops to assign rows and cols whereas vectorized methods may be available. — Parfait
– Parfait, Commented May 10, 2019 at 18:52
@Wimpel Unfortunately, I cannot use matrices to initialize because I need to preserve the various classes of the "old_df" dataframe. — jvel
– jvel, Commented May 10, 2019 at 19:10
@Parfait I am trying to initialize the number of rows for the exact reason that I don't want to loop through anything. I've stated the "why" in the question (I need to rbind these two dataframes). — jvel
– jvel, Commented May 10, 2019 at 19:10
I'm confused. Then why do you need an empty data frame? Please show us a sample of both data frames. — Parfait
– Parfait, Commented May 10, 2019 at 19:26

Rui Barradas · Accepted Answer · 2019-05-10 18:58:50Z

2

If I understand the question correctly, the following hack will do what the OP wants. It creates the number of rows by setting the row.names attribute directly. And if a dataframe has row names, it must have the corresponding rows.

empty_df <- old_df[0, ]
attr(empty_df, 'row.names') <- 1:nrow(new_df)

str(empty_df)
#'data.frame':  300 obs. of  5 variables:
# $ Sepal.Length: num 
# $ Sepal.Width : num 
# $ Petal.Length: num 
# $ Petal.Width : num 
# $ Species     : Factor w/ 3 levels "setosa","versicolor",..:

The dataframe empty_df now has 300 rows.

Data creation code.

The test data creation code uses the built-in dataset iris.

set.seed(1234)

old_df <- iris
new_df <- rbind(iris, iris)
new_df <- new_df[, sample(ncol(new_df))]

answered May 10, 2019 at 18:58

Rui Barradas

78k8 gold badges41 silver badges75 bronze badges

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

1 Comment

jvel Over a year ago

Perfect! Thank you! That is exactly what I need.

Collectives™ on Stack Overflow

How to create an empty dataframe with specified columns AND number of rows in R?

1 Answer 1

1 Comment

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

1 Comment

Your Answer

Sign up or log in

Post as a guest

Related