plot_roc: Plot ROC curves
In nlanderson9/languagePredictR: Predict Outcomes from Natural Language

Description Usage Arguments Value See Also Examples

This function plots ROC curves from the results of the assess_models function

plot_roc(
  ...,
  individual_plot = TRUE,
  combined_plot = TRUE,
  facet_plot = TRUE,
  facet_summary = TRUE,
  colors,
  model_names,
  plot_auc_polygon = TRUE,
  plot_ci = TRUE,
  line_size = 1,
  print_auc = TRUE,
  print_ci = TRUE,
  print_auc_ci_font_size = 4,
  print_auc_ci_x,
  print_auc_ci_y,
  plot_legend = TRUE,
  plot_title,
  facet_n_row = NULL,
  facet_n_col = 2,
  legend_spacing = FALSE
)

`...`	Output(s) of the `language_model`, `comparison_model`, or `test_language_model` functions
`individual_plot`	If TRUE, graphs individual ROC curves for each model. Defaults to TRUE.
`combined_plot`	If TRUE, and modelAssessment contains multiple models, graphs a plot with all ROC curves overlapping. Defaults to TRUE.
`facet_plot`	If TRUE, and modelAssessment contains multiple models, graphs a faceted plot with all ROC curves included. Defaults to TRUE.
`facet_summary`	If TRUE, and modelAssessment contains multiple models, the facet_plot will include a plot with all ROC curves overlapping. Defaults to TRUE.
`colors`	A vector of colors to use for each model's ROC curve.
`model_names`	A vector of strings to use as titles/names for each model.
`plot_auc_polygon`	If TRUE, the area below with ROC curve with the lowest AUC will be shaded in. Defaults to TRUE.
`plot_ci`	If TRUE, a confidence band will be plotted around each ROC curve. Defaults to TRUE.
`line_size`	A numeric representing the width of the ROC curve line. Defaults to 1.
`print_auc`	If TRUE, the value of the AUC will be printed on the plot. Defaults to TRUE.
`print_ci`	If TRUE, the range of the confidence interval will be printed on the plot. Defaults to TRUE.
`print_auc_ci_font_size`	The font size for printed values for the AUC and confidence interval. Defaults to 4.
`print_auc_ci_x`	A vector of x (horizontal) positions determining where on the plot the AUC and confidence interval values will be printed.
`print_auc_ci_y`	A vector of y (vertical) positions determining where on the plot the AUC and confidence interval values will be printed.
`plot_legend`	If TRUE, a legend will be printed on all plots.
`plot_title`	The title of the plot
`facet_n_row`	The number of rows used to plot the facet_plot. Defaults to NULL.
`facet_n_col`	The number of columns used to plot the facet_plot. Defaults to 2.
`legend_spacing`	If TRUE, there will be spacing between the legend items. Defaults to FALSE.

Nothing (this function plots a series of graphs)

language_model, comparison_model, test_language_model

## Not run: 
strong_movie_review_data$cleanText = clean_text(strong_movie_review_data$text)
mild_movie_review_data$cleanText = clean_text(mild_movie_review_data$text)

# Using language to predict "Positive" vs. "Negative" reviews
# Only for strong reviews (ratings of 1 or 10)
movie_model_strong = language_model(strong_movie_review_data,
                                     outcome = "valence",
                                     outcomeType = "binary",
                                     text = "cleanText",
                                     progressBar = FALSE)

# Using language to predict "Positive" vs. "Negative" reviews
# Only for mild reviews (ratings of 4 or 7)
movie_model_mild = language_model(mild_movie_review_data,
                                     outcome = "valence",
                                     outcomeType = "binary",
                                     text = "cleanText",
                                     progressBar = FALSE)


# Plot ROC curves
plot_roc(movie_model_strong, movie_model_mild)

## End(Not run)