A Statistically Sound 'data.frame' Processor/Conditioner


[Up] [Top]

Documentation for package ‘vtreat’ version 1.0.3

Help Pages

vtreat-package vtreat: A Statistically Sound 'data.frame' Processor/Conditioner
buildEvalSets Build set carve-up for out-of sample evaluation.
catScore return significnace 1 variable logistic regression
designTreatmentsC Build all treatments for a data frame to predict a categorical outcome.
designTreatmentsN build all treatments for a data frame to predict a numeric outcome
designTreatmentsZ Design variable treatments with no outcome variable.
format.vtreatment Display treatment plan.
getSplitPlanAppLabels read application labels off a split plan.
kWayCrossValidation k-fold cross validation, a splitFunction in the sense of vtreat::buildEvalSets
kWayStratifiedY k-fold cross validation stratified on y, a splitFunction in the sense of vtreat::buildEvalSets
kWayStratifiedYReplace k-fold cross validation stratified with replacement on y, a splitFunction in the sense of vtreat::buildEvalSets .
linScore Return in-sample linear stats and scaling.
makekWayCrossValidationGroupedByColumn Build a k-fold cross validation splitter, respecting (never splitting) groupingColumn.
mkCrossFrameCExperiment Run categorical cross-frame experiment.
mkCrossFrameNExperiment Run a numeric cross frame experiment.
novel_value_summary Report new/novel appearances of character values.
oneWayHoldout One way holdout, a splitFunction in the sense of vtreat::buildEvalSets.
prepare Apply treatments and restrict to useful variables.
print.vtreatment Print treatmentplan.
problemAppPlan check if appPlan is a good carve-up of 1:nRows into nSplits groups
track_values Track unique character values for variables.
vnames New treated variable names from a treatmentplan$treatment item.
vorig Original variable name from a treatmentplan$treatment item.
vtreat vtreat: A Statistically Sound 'data.frame' Processor/Conditioner