How to replace the maximum value in each row in a dataframe by an individual value-CodePudding

Today I failed in doing a simple task:

I want to replace the maximum value in each row of a data frame with a specific individual value that I have computed before. I did not know how I could tell

DF <- structure(list(age_n = c(2, 3, 3, 3, 3, 3), female_noe_n = c(0.674555812063115, 
0.681331945419623, 0.632659475334461, 0.649517172953091, 0.64918662986253, 
0.667531771388686), male_noe_n = c(0.734797290317925, 0.742565052946117, 
0.715791062610646, 0.729425965096302, 0.710171830071103, 0.734962561863205
), female_v_n = c(0.745209397670608, 0.86387436718214, 0.915191181991792, 
1.01146185711779, 1.04658206048994, 1.11211222819373), male_v_n = c(0.783635031948367, 
0.860486300503885, 0.909571949452249, 0.995595788770845, 1.03071599214299, 
1.12996155508404)), row.names = c(NA, 6L), class = "data.frame")

ID <- apply(DF, 1, function(x) which.max(x))

# Now I need to overwrite the rowwise maximum by ceiling 
overwrite_value <- 
  sapply(1:nrow(DF), function(x)
    DF[x,ID[[x]]] %>% ceiling())

I guess there are simple solutions and I am stuck in something too complex

CodePudding user response：

With row/column indexing it will be more efficient i.e. get the column index per row for the max value with max.col, cbind with the row sequence to return a matrix of row/column index, extract the value, get the ceiling and assign

m1 <- cbind(seq_len(nrow(DF)), max.col(DF[-1]))
DF[m1] <- ceiling(DF[m1])

CodePudding user response：

Here is a dplyr solution:

library(dplyr)

individual_value <- 5

DF %>% 
  rowwise() %>% 
  mutate( x = max(c_across(2:5))) %>% 
  mutate(across(2:5, ~case_when(. == x ~ individual_value,
                                TRUE ~ .)), .keep="unused"
         )

  age_n female_noe_n male_noe_n female_v_n male_v_n
  <dbl>        <dbl>      <dbl>      <dbl>    <dbl>
1     2        0.675      0.735      0.745    5    
2     3        0.681      0.743      5        0.860
3     3        0.633      0.716      5        0.910
4     3        0.650      0.729      5        0.996
5     3        0.649      0.710      5        1.03 
6     3        0.668      0.735      1.11     5