R6 Class for rtemis Supervised Models
rtModAn object of class R6ClassGenerator of length 24.
mod.nameLearner algorithm name
y.trainTraining y data
y.testTesting y data
x.nameName of x data
y.nameName of y data
xnamesCharacter vector: Column names of x
modTrained model
typeClassification, Regression, or Survival
parametersList of hyperparameters used when building model
fittedFitted values
se.fitStandard error of the fit
error.trainTraining error
predictedPredicted values
se.predictionStandard error of the prediction
error.testTesting error
questionQuestion the model is hoping to answer
extraAlgorithm-specific output
sessionInfoThe output of sessionInfo() at the time the model was trained
Note on terminology: The training set ('x.train') is used to build a model and calculate training error (rtMod$error.train). The testing set ('x.test') is not seen during training and only used to create predictions (rtMod$predicted) using the trained model and calculate error metrics (rtMod$error.test). This reflects generalizability of the model and is the error we care the most about. The validation set is used during model tuning. Within rtemis, validation sets are created automatically by resample when appropriate, they are not generally input by the user, with few exceptions, as documented in individual functions.