I am using R and have run a Gurobi optimization using model$vtype <- "B" where all elements of result$x are numeric and binary
For some reason one of my result variables that supposedly equals 1, does not equal 1 when I test for it. See the following from the console:
Browse[2]> result$x
[1] 0 0 0 0 0 0 0 0 0 1 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0
[90] 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
[179] 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0
Browse[2]> result$x[15]
[1] 1
Browse[2]> result$x[15]==1
[1] FALSE
All the 8 other elements of result$x that equal 1 successfully equal TRUE when I test whether they equal 1, it's only the 15th element that is causing an issue.
Any ideas why this is occurring? I'm using a binary variable type, so I don't get how 1 != 1
Other possibly relevant info:
Browse[2]> is.numeric(result$x[15])
[1] TRUE
Browse[2]> typeof(result$x[15])
[1] "double"
Browse[2]> which(result$x==1)
[1] 10 33 48 81 127 133 151 198
Further info requested:
Browse[2]> typeof(result$x)
[1] "double"
When I cast it to integer, the 15th element for some reason becomes 0
instead of 1.. can anyone explain why this is happening and how I should properly adjust it? See below:
as.integer(result$x)
[1] 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0
[90] 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
[179] 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0
Thanks