I’m trying to appply a function using mapping. However, i’m getting “Caused by error in qr.default():
! NA/NaN/Inf in foreign function call (arg 1)” error. There are no NA in the data.

My code:

exp_fit <- function(df){
  nls(capacity ~ a*(1 + b)**(year-1995),
      data = df,
      start = list(a = 100,
                   b = 0.4))
}
regions.bp = tibble(name = c("EU", "Global"),
                    list = list("EU",
                                "Global"),
                    span = list(1995:2010,
                                1995:2010))

time.slice.length <- 5:10

 for (r in 1:dim(regions.bp)[1]) {
   time.span <- regions.bp$span[[r]]
   for (t in time.slice.length){
     # Get individual time slices for each length t
     for (s in 1:(length(time.span)-t+1)){
       slice <- time.span[s:(s+t-1)]
       temp <- data.bp %>% 
         filter(region %in% regions.bp$list[[r]],
                year %in% slice) %>% 
         group_by(technology, year) %>% 
         summarise(capacity = sum(capacity)) %>% 
         group_by(technology) %>% 
         nest() %>% 
         mutate(model.exp = map(data, exp_fit))
     }
  }
}

Error in mutate():
In argument: model.exp = map(data, exp_fit).
In group 1: technology = "geothermal".
Caused by error in map():
In index: 1.
Caused by error in qr.default():
NA/NaN/Inf in foreign function call (arg 1)

I tried to omit NA values and tried a loop instead of map function. However it returns the same error.