constructModels: Construct ensemble models from a data.frame of optional...
In jknowles/ModelEWS: Wisconsin Dropout Early Warning System

Description Usage Arguments Details Value

Construct ensemble models from a data.frame of optional models

1 2	constructModels(modelResultsName, grade, mode, DV = "grad_ind", nmodels = 8, conn)

`modelResultsName`	a prefix that includes the path and filename of the file on the disk where the model search results object is stored
`grade`	a numeric representing the grade level of the object
`mode`	a character representing whether the model was run in dev or in production mode
`DV`	a character representing the name of the dependent variable in the model
`nmodels`	the number of models to include in an ensemble
`conn`	a connection string pointing to the database where the gatherData function can extract data from for model building

By default the function builds four kinds of models. It builds ensembles based on the top N performing models, and on the N most dissimilar models. To build these it uses both the caretStack and caretEnsemble method.

Nothing. Writes the model objects to disk

jknowles/ModelEWS documentation built on May 19, 2019, 11:42 a.m.