is.na10 is a helper function for creating heatmaps to diagnose missing value patterns. It is similar to is.na but instead of returning a logical TRUE/FALSE vector (or matrix) it returns a numeric 1/0 output. This enables the heatmaply function to be used on the data.
is.na10(x, ...)a vector, matrix or data.frame.
not used.
Returns a numeric (instead of a logical) variable/matrix of 1 (missing) or 0 (not missing) values (hence the name is.na10) while still preserving the attributes resulted from running is.na.
These are useful for funnelling into a heatmap (see the examples).
if (FALSE) {
x <- mtcars
x <- data.frame(x)
x$am <- factor(x$am)
x$vs <- factor(x$vs)
set.seed(2017 - 01 - 19)
x[sample(nrow(x))[1:6], sample(ncol(x))[1:6]] <- NA
# nice grey colors from here: https://github.com/njtierney/visdat/blob/master/R/vis_miss_ly.R
x %>%
is.na10() %>%
heatmaply(colors = c("grey80", "grey20"), dendrogram = "none")
x %>%
is.na10() %>%
heatmaply(colors = c("grey80", "grey20"), k_col = 2, k_row = 2)
heatmaply(is.na10(airquality),
grid_gap = 1,
colors = c("grey80", "grey20"), k_col = 2, k_row = 2
)
}