ml_bwplot: A function to visualize the model performance.

Description Usage Arguments Details See Also Examples

Description

A wrapper around the bwplot function in caret to visualize models even if models are trained using different metrics.

Usage

1
ml_bwplot(models, metric = NULL)

Arguments

models

A list of models from ml_list function or manually combine finely tuned models from ml_tune or train function from caret package.

metric

A character, the metric of model performance, the available values are "ROC","Sens","Spec","Accuracy","Kappa".

Details

If the models contains metrics from c("ROC","Sens","Spec") and c("Accuracy","Kappa"), then it will plot two graphs. You could not plot the "ROC","Sens" or "Spec" value for a model if the model trained by using "Accuracy" or "Kappa", and vice versa.

See Also

To see how the list of models are generated ml_list and ml_tune. The bwplot function is not documented on CRAN but you could see it on caret tutorial https://topepo.github.io/caret/model-training-and-tuning.html

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
# plot all the metrics
## Not run: 
  ml_bwplot(models)

## End(Not run)

# plot only the ROC metric for all models.
## Not run: 
ml_bwplot(models,metric="ROC")

## End(Not run)

edwardcooper/automl documentation built on June 3, 2019, 1:05 a.m.