Delete all rows with non-missing values in certain columns

[Peter_Griffin]

I have some data.frames with different number of columns. For each such data.frame, I need to delete all rows with non-missing values in columns 2:ncol(data.frame)-1, which means all the columns except for the first and last columns. How could I do that?

[woodward]

How about drop_na() from tidyverse? You can put the columns to exclude in the brackets.

df <- df %>%
       drop_na(-first_column_name, -last_column_name)

[andresrcs]

Could you please ask your questions with a minimal REPRoducible EXample (reprex)?

I know this question is not that complicated but It could bring speculation and unnecessary back and forths

[Peter_Griffin]

Thanks. My task is computationally intensive, do we have a data.table version of drop_na?

[woodward]

If you need to subset a big dataframe it's often fast to do it like this:

keeps <- rowSums(is.na(df)) == 0 # identify rows to keep
df <- df[keeps, ] # keep them