This function plots a whole dashboard with a model's results. It will automatically detect if it's a categorical or regression's model by checking how many different unique values the independent variable (tag) has.
mplot_full(tag, score, multis = NA, splits = 8, thresh = 6,
subtitle = NA, model_name = NA, plot = TRUE, save = FALSE,
subdir = NA, file_name = "viz_full.png")Vector. Real known label
Vector. Predicted value or model's result. Must be numeric for categorical binomial models and continuous regression models; must be categorical for multi-categorical models (also need multis param).
Data.frame. Containing columns with each category score (only used when more than 2 categories coexist)
Integer. Numer of separations to plot
Integer. Threshold for selecting binary or regression models: this number is the threshold of unique values we should have in 'tag' (more than: regression; less than: classification)
Character. Subitle to show in plot
Character. Model's name
Boolean. Plot results? If not, plot grid object returned
Boolean. Save output plot into working directory
Character. Sub directory on which you wish to save the plot
Character. File name as you wish to save the plot
Other Machine Learning: ROC,
clusterKmeans, conf_mat,
export_results, gain_lift,
h2o_automl, h2o_predict_API,
h2o_predict_MOJO,
h2o_predict_binary,
h2o_predict_model,
h2o_selectmodel, impute,
iter_seeds, model_metrics,
mplot_conf, mplot_cuts_error,
mplot_cuts, mplot_density,
mplot_gain, mplot_importance,
mplot_lineal, mplot_metrics,
mplot_response, mplot_roc,
mplot_splits, msplit
Other Visualization: corr_plot,
distr, freqs_df,
freqs, mplot_conf,
mplot_cuts_error, mplot_cuts,
mplot_density, mplot_gain,
mplot_importance,
mplot_lineal, mplot_metrics,
mplot_response, mplot_roc,
mplot_splits, noPlot,
plot_survey, theme_lares2,
theme_lares, tree_var