Plug an existing scikit learn python model into the PLP framework

Usage

createSklearnModel(
  modelLocation = "/model",
  covariateMap = data.frame(columnId = 1:2, covariateId = c(1, 2), ),
  covariateSettings,
  populationSettings,
  isPickle = TRUE
)

Arguments

modelLocation

The location of the folder that contains the model as model.pkl

covariateMap

A data.frame with the columns: columnId and covariateId. covariateId from FeatureExtraction is the standard OHDSI covariateId. columnId is the column location the model expects that covariate to be in. For example, if you had a column called 'age' in your model and this was the 3rd column when fitting the model, then the values for columnId would be 3, covariateId would be 1002 (the covariateId for age in years) and

covariateSettings

The settings for the standardized covariates

populationSettings

The settings for the population, this includes the time-at-risk settings and inclusion criteria.

isPickle

If the model should be saved as a pickle set this to TRUE if it should be saved as json set this to FALSE.

Value

An object of class plpModel, this is a list that contains: model (the location of the model.pkl), preprocessing (settings for mapping the covariateIds to the model column mames), modelDesign (specification of the model design), trainDetails (information about the model fitting) and covariateImportance.

You can use the output as an input in PatientLevelPrediction::predictPlp to apply the model and calculate the risk for patients.

Details

This function lets users add an existing scikit learn model that is saved as model.pkl into PLP format. covariateMap is a mapping between standard covariateIds and the model columns. The user also needs to specify the covariate settings and population settings as these are used to determine the standard PLP model design.