mplot_full: MPLOTS Score Full Report Plots

In laresbernardo/lares: Analytics & Machine Learning Sidekick

MPLOTS Score Full Report Plots

Description

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 dependent variable (tag) has.

Usage

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"
)

Arguments

tag	Vector. Real known label.
score	Vector. Predicted value or model's result.
multis	Data.frame. Containing columns with each category probability or score (only used when more than 2 categories coexist).
splits	Integer. Number of separations to plot
thresh	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)
subtitle	Character. Subtitle to show in plot
model_name	Character. Model's name
plot	Boolean. Plot results? If not, plot grid object returned
save	Boolean. Save output plot into working directory
subdir	Character. Sub directory on which you wish to save the plot
file_name	Character. File name as you wish to save the plot

Value

Multiple plots gathered into one, showing tag vs score performance results.

See Also

Other ML Visualization: mplot_conf(), mplot_cuts(), mplot_cuts_error(), mplot_density(), mplot_gain(), mplot_importance(), mplot_lineal(), mplot_metrics(), mplot_response(), mplot_roc(), mplot_splits(), mplot_topcats()

Examples

Sys.unsetenv("LARES_FONT") # Temporal
data(dfr) # Results for AutoML Predictions
lapply(dfr, head)

# Dasboard for Binomial Model
mplot_full(dfr$class2$tag, dfr$class2$scores,
  model_name = "Titanic Survived Model"
)

# Dasboard for Multi-Categorical Model
mplot_full(dfr$class3$tag, dfr$class3$score,
  multis = subset(dfr$class3, select = -c(tag, score)),
  model_name = "Titanic Class Model"
)

# Dasboard for Regression Model
mplot_full(dfr$regr$tag, dfr$regr$score,
  model_name = "Titanic Fare Model"
)

laresbernardo/lares documentation built on Oct. 23, 2024, 12:05 p.m.