Plot a frequency chart to compare actual and predicted values.
frequency_plot(count_labels, actual, pred, colours = character(0))an object from class "trellis"
character, labels to be used.
numeric, the observed probabilities for the different count
specified in count_labels.
data.frame of predicted values. Should have the same number of rows as actual and one column per model. The columns' names will be used as labels for the different models.
character vector of colour codes with length
ncol(pred) + 2. If colours is missing or
length(colours) < ncol(pred) + 2, the remaining colours are generated
using RColorBrewer::brewer.pal.
In order to compare actual and fitted values, a barchart plot is created. It is the user's responsibility to provide the count, observed and fitted values.
If argument colour is missing or not of sufficient length, the colours
are set automatically using a function from package RColorBrewer.
The bar chart is created with lattice::barchart. If
frequency_plot is called from the command line, the returned value is
automatically ‘printed’ (i.e., the plot is produced). Otherwise, for
example in scripts, you may need to use print() on the returned value.