suppress non informative output in ml chapter

[Daenarys8]

ping #635

[TuomasBorman]

We use ROC plots that are not ggplot. Can you switch to ggplot? Use for instance plotROC package. Remember to update DSCRIPTION file

[TuomasBorman]

And try this

library(xgboost)
library(miaViz)

# Get feature importance
df <- xgb.importance(model = model$finalModel) |> as.data.frame()
rownames(df) <- df[["Feature"]]
df <- df[, c("Gain"), drop = FALSE]
df <- as.matrix(df)

# Create a plot
p <- plotLoadings(df, ncomponents = 1, n = 20, show.color = FALSE)
p

[TuomasBorman]

I removed unused ROCR from DESCRIPTION. Moreover, if you add packages, you have to bump the version. Otherwise the environment is not updated