R Error: got class "character", should be or extend class "numericOrNA"

[omario]

I am working with the R programming language. I am trying to use the "genetic algorithm" to optimize the following function ("f") using the "GA" library:

# create data for example

num_var_1 <- abs(rnorm(1000, 10, 1))
num_var_2 <- abs(rnorm(1000, 10, 5))
num_var_3 <- abs(rnorm(1000, 10, 10))
num_var_4 <- abs(rnorm(1000, 10, 10))
num_var_5 <- abs(rnorm(1000, 10, 10))

factor_1 <- c("0","B", "C")
factor_2 <- c("0","BB", "CC")
factor_3 <- c("0","BBB", "CCC", "DDD")
factor_4 <- c("0","BBBB", "CCCC", "DDDD", "EEEE")
factor_5 <- c("0","BBBBB", "CCCCC", "DDDDD", "EEEEE", "FFFFFF")

factor_var_1 <- as.factor(sample(factor_1, 1000, replace=TRUE, prob=c(0.3, 0.5, 0.2)))
factor_var_2 <-  as.factor(sample(factor_2, 1000, replace=TRUE, prob=c(0.5, 0.3, 0.2)))
factor_var_3 <-  as.factor(sample(factor_3, 1000, replace=TRUE, prob=c(0.5, 0.2, 0.2, 0.1)))
factor_var_4 <-  as.factor(sample(factor_4, 1000, replace=TRUE, prob=c(0.5, 0.2, 0.1, 0.1, 0.1)))
factor_var_5 <-  as.factor(sample(factor_4, 1000, replace=TRUE, prob=c(0.3, 0.2, 0.1, 0.1, 0.1)))

id = 1:1000

my_data = data.frame(id,num_var_1, num_var_2, num_var_3, num_var_4, num_var_5, factor_var_1, factor_var_2, factor_var_3, factor_var_4, factor_var_5)

#randomly add some zeros to the data
my_data[] <- lapply(my_data, function(x) {
  x[sample(seq_along(x), length(x)/2)] <- 0
  x
})

#load libraries
library(dplyr)
library(GA)

#define function for optimization
 
f <- function(x1, x2, x3, x4, x5, x8) {

target1 = x1
target2 = x2
target3 = x3
target4 = x4
target5 = x5
target8 = x8

result <-  filter(my_data, num_var_1 < target1 & num_var_2 < target2 & num_var_3 < target3 & num_var_4 <target4 & num_var_5 <target5 & factor_var_3 %in% target8 )

return(sum(result == '0', na.rm = TRUE) / prod(dim(result)) * 100)

}

# run optimization

GA <- ga(type = "real-valued", 
         fitness = function(x)  f(x[1], x[2], x[3], x[4], x[5], x[6]),
         lower = c( min(num_var_1),  min(num_var_2),  min(num_var_3),  min(num_var_4),  min(num_var_5), c("0","BBB", "CCC", "DDD")), upper = c( max(num_var_1),  max(num_var_2),  max(num_var_3),  max(num_var_4),  max(num_var_5),  c("0","BBB", "CCC", "DDD")), 
         popSize = 50, maxiter = 1000, run = 100)

Problem: However, this returns the following error:

Error in validObject(.Object) : 
  invalid class “ga” object: 1: invalid object for slot "lower" in class "ga": got class "character", should be or extend class "numericOrNA"
invalid class “ga” object: 2: invalid object for slot "upper" in class "ga": got class "character", should be or extend class "numericOrNA"

I think this error is because the x[6] argument takes "factors" as input, and not "numeric" input. Without the x[6] argument, the optimization function works fine:

f_mod <- function(x1, x2, x3, x4, x5) {
    
    target1 = x1
    target2 = x2
    target3 = x3
    target4 = x4
    target5 = x5
    
    
    result <-  filter(my_data, num_var_1 < target1 & num_var_2 < target2 & num_var_3 < target3 & num_var_4 <target4 & num_var_5 <target5 )
    
    return(sum(result == '0', na.rm = TRUE) / prod(dim(result)) * 100)
    
}


GA <- ga(type = "real-valued", 
         fitness = function(x)  f_mod(x[1], x[2], x[3], x[4], x[5]),
         lower = c( min(num_var_1),  min(num_var_2),  min(num_var_3),  min(num_var_4),  min(num_var_5)), upper = c( max(num_var_1),  max(num_var_2),  max(num_var_3),  max(num_var_4),  max(num_var_5)), 
         popSize = 50, maxiter = 100, run = 100)

My Question: Can someone please show me how to fix this problem? Are there any optimization algorithms/packages in R that can optimize functions having both "numeric" and "factor" inputs?

Thanks!

[valentingar]

Hi there,
not having used that package myself, I do think you are misunderstanding the error message. The problem lies within the lower and upper bounds you have defined. Not only is the length not matching the number of arguments that you define in you first example (note that c(1,c(2,3)) = c(1,2,3)), but you are also providing character values (c("BBB", "CCC," "DDD")), where numeric or NA was expected. I don't think you can really use such optimisations for categorical variables? If these values you provided (c("0", "BBB", "CCC", "DDD")) are the only expected values in your function, I would suggest optimising only the five numeric variables and repeating this for the discrete values of x6.
Hope this helps.
Best
Valentin