saveData throws NAs in train_data

[cebarboza]

Hi,

I noticed there was a change in saveData() compared to the version I was using in tests. If data is TRUE or FALSE, is not converting the data to 0 and 1 like before in here.

train_data <- data.frame(check.names = FALSE,
                          outcomeCount = c(FALSE,FALSE,FALSE,
                                           FALSE,FALSE,TRUE),
                           `198124209` = c(FALSE,FALSE,FALSE,
                                           FALSE,FALSE,TRUE),
                           `316139209` = c(FALSE,FALSE,FALSE,
                                           FALSE,FALSE,FALSE),
                           `316139210` = c(FALSE,FALSE,FALSE,
                                           FALSE,FALSE,FALSE)
                        )

binary_cols <- sapply(1:ncol(train_data), function(c) all(train_data[[c]] %in% 0:1))
train_data[binary_cols] <- lapply(colnames(train_data[binary_cols]), function(c) factor(train_data[[c]], levels=c("0","1"), labels=c(0,1)))
  
print(train_data[binary_cols])
#>   outcomeCount 198124209 316139209 316139210
#> 1         <NA>      <NA>      <NA>      <NA>
#> 2         <NA>      <NA>      <NA>      <NA>
#> 3         <NA>      <NA>      <NA>      <NA>
#> 4         <NA>      <NA>      <NA>      <NA>
#> 5         <NA>      <NA>      <NA>      <NA>
#> 6         <NA>      <NA>      <NA>      <NA>

Maybe something like to account for TRUE and FALSE?

train_data <- data.frame(check.names = FALSE,
                         outcomeCount = c(FALSE,FALSE,FALSE,
                                          FALSE,FALSE,TRUE),
                         `198124209` = c(FALSE,FALSE,FALSE,
                                         FALSE,FALSE,TRUE),
                         `316139209` = c(FALSE,FALSE,FALSE,
                                         FALSE,FALSE,FALSE),
                         `316139210` = c(FALSE,FALSE,FALSE,
                                         FALSE,FALSE,FALSE)
)

binary_cols <- sapply(train_data, function(col) all(col %in% c(0, 1, TRUE, FALSE)))

# Convert TRUE/FALSE to 1/0 and create factors
train_data[binary_cols] <- lapply(train_data[binary_cols], function(col) {
  col <- as.numeric(as.logical(col))  # Convert TRUE/FALSE to 1/0
  factor(col, levels = c(0, 1), labels = c(0, 1))  # Convert to factors
})

print(train_data[binary_cols])
#>   outcomeCount 198124209 316139209 316139210
#> 1            0         0         0         0
#> 2            0         0         0         0
#> 3            0         0         0         0
#> 4            0         0         0         0
#> 5            0         0         0         0
#> 6            1         1         0         0

^{Created on 2024-09-30 with reprex v2.1.1}