An S3 object generated by evalmod. The autoplot function accepts the following codeS3 objects for two different modes, 'rocprc' and 'basic'.

1. ROC and Precision-Recall curves (mode = 'rocprc')

   S3 object	# of models	# of test datasets
   sscurves	single	single
   mscurves	multiple	single
   smcurves	single	multiple
   mmcurves	multiple	multiple

2. Basic evaluation measures (mode = 'basic')

   S3 object	# of models	# of test datasets
   sspoints	single	single
   mspoints	multiple	single
   smpoints	single	multiple
   mmpoints	multiple	multiple

See the Value section of evalmod for more details.

A character vector with the following curve types.

1. ROC and Precision-Recall curves (mode = 'rocprc')

   curvetype	description
   ROC	ROC curve
   PRC	Precision-Recall curve

   Multiple curvetype can be combined, such as c('ROC', 'PRC').

2. Basic evaluation measures (mode = 'basic')

   curvetype	description
   error	Normalized ranks vs. error rate
   accuracy	Normalized ranks vs. accuracy
   specificity	Normalized ranks vs. specificity
   sensitivity	Normalized ranks vs. sensitivity
   precision	Normalized ranks vs. precision
   mcc	Normalized ranks vs. Matthews correlation coefficient
   fscore	Normalized ranks vs. F-score

   Multiple curvetype can be combined, such as c('precision', 'sensitivity').

Following additional arguments can be specified.

type

A character to specify the line type as follows.

'l'

lines

'p'

points

'b'

both lines and points

show_cb

A Boolean value to specify whether point-wise confidence bounds are drawn. It is effective only when calc_avg of the evalmod function is set to TRUE .

raw_curves

A Boolean value to specify whether raw curves are shown instead of the average curve. It is effective only when raw_curves of the evalmod function is set to TRUE.

show_legend

A Boolean value to specify whether the legend is shown.

ret_grob

A logical value to indicate whether autoplot returns a grob object. The grob object is internally generated by arrangeGrob. The grid.draw function takes a grob object and shows a plot. It is effective only when a multiple-panel plot is generated, for example, when curvetype is c('ROC', 'PRC').

reduce_points

A Boolean value to decide whether the points should be reduced when mode = 'rocprc'. The points are reduced according to x_bins of the evalmod function. The default values is TRUE.