which() does not work for some numbers

Asked Jun 28 '18 at 10:47

Active Jun 28 '18 at 10:51

Viewed 42 times

I have a dataframe where the first column consists of the numbers 0,0.01,...,0.99,1. Using the which() function i'm trying to find the row indices that contain a certain set of numbers in the first three columns. Peculiarly, this does does not work for the numbers 0.35, 0.41, 0.47, 0.57, 0.69, 0.70, 0.82, 0.83, 0.94 and 0.95 in the first column. I'm sure they exist in the data frame, but which() returns integer(0) each time. Can anybody help?

StateIndex = function(state)
{
  state = as.numeric(state)
  aux = which(states[,colnames(states)[1]] == state[1])
  for (i in 2:3)
  {
    aux = aux[which(states[aux,colnames(states)[i]] == state[i])]
  }
return(aux)
}

states = list()
states[[1]] = seq(0,1,by=1/M) #zeta1 at last action
states[[2]] = 1:(m+1)         #i (last observed det. level)
states[[3]] = 0:N             #j (time since last action)

states = expand.grid(states)

nstates = dim(states)[1]

edited Jun 28 '18 at 10:51

asked Jun 28 '18 at 10:47

Rob Hermsen

are you sure it's exactly `0.35`, not, e.g., `0.3500000000000001`? – lebatsnok Jun 28 '18 at 10:50
Yes, i added the data frame for clarity. The first column contains the numbers 0 to 1 with steps of 1/100. – Rob Hermsen Jun 28 '18 at 10:52
search on stackoverfelow for question first then ask. – MRizwan33 Jun 28 '18 at 10:56
I did. I already tried isTRUE(all.equal()). Sorry I did not mention that. – Rob Hermsen Jun 28 '18 at 11:03
1

Can you clarify your data set? What are `M`, `m`, and `N`? Ideally, provide code and data that can simply be copy and pasted to reproduce your error. – Dan Jun 28 '18 at 11:12
Try `sprintf('%.100f', states$Var1[23313])` and you will see it's not 0.82 :) – konvas Jun 28 '18 at 11:13

which() does not work for some numbers

0 Answers0