How to apply function to each row of dataframe in R?

Question

I want to apply given function to each row of dataframe and use another values of row, as input parameters/arguments:

Model <- c("H5", "H5", "H5","H4")
Length <- c(6, 6, 6, 6)
Code <- c("030299", "010121","030448","030324")

df <- data.frame(Model,Length,Code)


Model   Length   Code
HS5       6     030299
HS5       6     010121
HS5       6     030448
HS4       6     030324

I want to apply the following code to each row and generate the outcome as a new column

Library(concordance)

concord(sourcevar = (each row of 'Code' column), origin = as.character(character in 'Model' column) , destination = "HS4", dest.digit = as.numeric(number in 'Length' column), all = F))

Documentation Page 6

Knowhow

Answer 1

You can use apply on the rows (MARGIN = 1).

apply(df, MARGIN = 1, function(x) concord(sourcevar = x[3], origin = x[1], destination = "HS4", dest.digit = x[2], all = F))

However, this does not work because there is no conversion dictionary between "HS4" and "HS4", so you can use apply only on the rows that are not HS4:

df$New <- df$Code
df[df$Model != "HS4", ]$New <- apply(df[df$Model != "HS4", ], 1, \(x) concord(sourcevar = x[colnames(df) == "Code"], 
                                               origin = x[colnames(df) == "Model"], destination = "HS4", 
                                               dest.digit = x[colnames(df) == "Length"], all = F))

  Model Length   Code    New
1   HS5      6 030299 030289
2   HS5      6 010121 010121
3   HS5      6 030448 030449
4   HS4      6 030324 030324

Daisy

These numbers (1, x[3], x[1] , x[3]) in the last code are number of columns, right? I am asking because I will need this code next time too, to apply another data – Knowhow

Yes they are. Exactly. – Unblessed

is it possible to change them by column names? instead of index – Knowhow

what this 1 means here, could you explain? "HS4", ], 1, (x) – Knowhow

This indicates you're applying the function on the rows (2 would columns) – Unblessed

The code runs too slow on big data, I think "apply" functions is root of of this problem. Is it possible to somehow modify last code to make it faster? I can open new question if you want – Knowhow

apply might be the problem indeed, but that might also be that your functions, when applied to a lot of rows, just take more time? I'd open a new question rather than edit this one bc it's a different problem. – Unblessed