CODEWITHSUNDEEP
r
nouse
nouse

Reputation: 3471

Subsetting a data.frame by a numerical vector

I have a dataframe with 358 rows

df <- data.frame(A=rnorm(358),B=rnorm(358))

and want to subset at these breaks:

breaks <- c(59,119,178,238,298)

resulting in six dataframes with

59, 60, 59, 60, 60, 60 rows.

(This is a result of a 6 x 60 sampling design, of which two samples have been lost.)

Is this possible?

Upvotes: 1

Views: 73

Answers (1)

Konrad
Konrad

Reputation: 18657

If you fancy a primitive loop solution you can simply iterate through you desired row numbers, I understand that you are interested in subsetting this data frame by row numbers.

set.seed(1)
df <- data.frame(A=rnorm(358),B=rnorm(358)) 
## Added 1
breaks <- c(1, 59,119,178,238,298, 358)

listDFs <- vector("list", length(breaks) - 1)

for (i in 1:(length(breaks)-1)) {
  # Take observations from this to that row
  listDFs[[i]] <- df[breaks[i]:breaks[i+1],]
}

Side point

I reckon that suggestion provided in comments by @Frank is better:

n = c(59, 60, 59, 60, 60, 60); split(df, rep(seq_along(n), n))

Loop would make sense if you are you looking for a visually convenient way to do more things on the generated object.

Results

> sapply(listDFs, dim)
     [,1] [,2] [,3] [,4] [,5] [,6]
[1,]   59   61   60   61   61   61
[2,]    2    2    2    2    2    2

Upvotes: 1

Related Questions