Specify settings for developing a single model

Specify settings for developing a single model

Usage

createModelDesign(
  targetId = NULL,
  outcomeId = NULL,
  restrictPlpDataSettings = createRestrictPlpDataSettings(),
  populationSettings = createStudyPopulationSettings(),
  covariateSettings = FeatureExtraction::createDefaultCovariateSettings(),
  featureEngineeringSettings = NULL,
  sampleSettings = NULL,
  preprocessSettings = NULL,
  modelSettings = NULL,
  splitSettings = createDefaultSplitSetting(),
  runCovariateSummary = TRUE
)

Arguments

targetId

The id of the target cohort that will be used for data extraction (e.g., the ATLAS id)

outcomeId

The id of the outcome that will be used for data extraction (e.g., the ATLAS id)

restrictPlpDataSettings

The settings specifying the extra restriction settings when extracting the data created using createRestrictPlpDataSettings().

populationSettings

The population settings specified by createStudyPopulationSettings()

covariateSettings

The covariate settings, this can be a list or a single 'covariateSetting' object.

featureEngineeringSettings

Either NULL or an object of class featureEngineeringSettings specifying any feature engineering used during model development

sampleSettings

Either NULL or an object of class sampleSettings with the over/under sampling settings used for model development

preprocessSettings

Either NULL or an object of class preprocessSettings created using createPreprocessingSettings()

modelSettings

The model settings such as setLassoLogisticRegression()

splitSettings

The train/validation/test splitting used by all analyses created using createDefaultSplitSetting()

runCovariateSummary

Whether to run the covariateSummary

Value

A list with analysis settings used to develop a single prediction model

Details

This specifies a single analysis for developing as single model

Examples

# L1 logistic regression model to predict the outcomeId 2 using the targetId 2
# with with default population, restrictPlp, split, and covariate settings
createModelDesign(
  targetId = 1,
  outcomeId = 2,
  modelSettings = setLassoLogisticRegression(seed=42),
  populationSettings = createStudyPopulationSettings(),
  restrictPlpDataSettings = createRestrictPlpDataSettings(),
  covariateSettings = FeatureExtraction::createDefaultCovariateSettings(),
  splitSettings = createDefaultSplitSetting(splitSeed = 42),
  runCovariateSummary = TRUE
)
#> $targetId
#> [1] 1
#> 
#> $outcomeId
#> [1] 2
#> 
#> $restrictPlpDataSettings
#> $studyStartDate
#> [1] ""
#> 
#> $studyEndDate
#> [1] ""
#> 
#> $firstExposureOnly
#> [1] FALSE
#> 
#> $washoutPeriod
#> [1] 0
#> 
#> $sampleSize
#> NULL
#> 
#> attr(,"class")
#> [1] "restrictPlpDataSettings"
#> 
#> $covariateSettings
#> $temporal
#> [1] FALSE
#> 
#> $temporalSequence
#> [1] FALSE
#> 
#> $DemographicsGender
#> [1] TRUE
#> 
#> $DemographicsAgeGroup
#> [1] TRUE
#> 
#> $DemographicsRace
#> [1] TRUE
#> 
#> $DemographicsEthnicity
#> [1] TRUE
#> 
#> $DemographicsIndexYear
#> [1] TRUE
#> 
#> $DemographicsIndexMonth
#> [1] TRUE
#> 
#> $ConditionGroupEraLongTerm
#> [1] TRUE
#> 
#> $ConditionGroupEraShortTerm
#> [1] TRUE
#> 
#> $DrugGroupEraLongTerm
#> [1] TRUE
#> 
#> $DrugGroupEraShortTerm
#> [1] TRUE
#> 
#> $DrugGroupEraOverlapping
#> [1] TRUE
#> 
#> $ProcedureOccurrenceLongTerm
#> [1] TRUE
#> 
#> $ProcedureOccurrenceShortTerm
#> [1] TRUE
#> 
#> $DeviceExposureLongTerm
#> [1] TRUE
#> 
#> $DeviceExposureShortTerm
#> [1] TRUE
#> 
#> $MeasurementLongTerm
#> [1] TRUE
#> 
#> $MeasurementShortTerm
#> [1] TRUE
#> 
#> $MeasurementRangeGroupLongTerm
#> [1] TRUE
#> 
#> $MeasurementRangeGroupShortTerm
#> [1] TRUE
#> 
#> $MeasurementValueAsConceptLongTerm
#> [1] TRUE
#> 
#> $MeasurementValueAsConceptShortTerm
#> [1] TRUE
#> 
#> $ObservationLongTerm
#> [1] TRUE
#> 
#> $ObservationShortTerm
#> [1] TRUE
#> 
#> $ObservationValueAsConceptLongTerm
#> [1] TRUE
#> 
#> $ObservationValueAsConceptShortTerm
#> [1] TRUE
#> 
#> $CharlsonIndex
#> [1] TRUE
#> 
#> $Dcsi
#> [1] TRUE
#> 
#> $Chads2
#> [1] TRUE
#> 
#> $Chads2Vasc
#> [1] TRUE
#> 
#> $includedCovariateConceptIds
#> logical(0)
#> 
#> $includedCovariateIds
#> logical(0)
#> 
#> $addDescendantsToInclude
#> [1] FALSE
#> 
#> $excludedCovariateConceptIds
#> logical(0)
#> 
#> $addDescendantsToExclude
#> [1] FALSE
#> 
#> $shortTermStartDays
#> [1] -30
#> 
#> $mediumTermStartDays
#> [1] -180
#> 
#> $endDays
#> [1] 0
#> 
#> $longTermStartDays
#> [1] -365
#> 
#> attr(,"fun")
#> [1] "getDbDefaultCovariateData"
#> attr(,"class")
#> [1] "covariateSettings"
#> 
#> $populationSettings
#> $binary
#> [1] TRUE
#> 
#> $includeAllOutcomes
#> [1] TRUE
#> 
#> $firstExposureOnly
#> [1] FALSE
#> 
#> $washoutPeriod
#> [1] 0
#> 
#> $removeSubjectsWithPriorOutcome
#> [1] TRUE
#> 
#> $priorOutcomeLookback
#> [1] 99999
#> 
#> $requireTimeAtRisk
#> [1] TRUE
#> 
#> $minTimeAtRisk
#> [1] 364
#> 
#> $riskWindowStart
#> [1] 1
#> 
#> $startAnchor
#> [1] "cohort start"
#> 
#> $riskWindowEnd
#> [1] 365
#> 
#> $endAnchor
#> [1] "cohort start"
#> 
#> $restrictTarToCohortEnd
#> [1] FALSE
#> 
#> attr(,"class")
#> [1] "populationSettings"
#> 
#> $sampleSettings
#> $sampleSettings[[1]]
#> $numberOutcomestoNonOutcomes
#> [1] 1
#> 
#> $sampleSeed
#> [1] 1
#> 
#> attr(,"fun")
#> [1] "sameData"
#> attr(,"class")
#> [1] "sampleSettings"
#> 
#> 
#> $featureEngineeringSettings
#> $featureEngineeringSettings[[1]]
#> list()
#> attr(,"fun")
#> [1] "sameData"
#> attr(,"class")
#> [1] "featureEngineeringSettings"
#> 
#> 
#> $preprocessSettings
#> $minFraction
#> [1] 0.001
#> 
#> $normalize
#> [1] TRUE
#> 
#> $removeRedundancy
#> [1] TRUE
#> 
#> attr(,"class")
#> [1] "preprocessSettings"
#> 
#> $modelSettings
#> $fitFunction
#> [1] "fitCyclopsModel"
#> 
#> $param
#> $param$priorParams
#> $param$priorParams$priorType
#> [1] "laplace"
#> 
#> $param$priorParams$forceIntercept
#> [1] FALSE
#> 
#> $param$priorParams$variance
#> [1] 0.01
#> 
#> $param$priorParams$exclude
#> [1] 0
#> 
#> 
#> $param$includeCovariateIds
#> NULL
#> 
#> $param$upperLimit
#> [1] 20
#> 
#> $param$lowerLimit
#> [1] 0.01
#> 
#> $param$priorCoefs
#> NULL
#> 
#> attr(,"settings")
#> attr(,"settings")$priorfunction
#> [1] "Cyclops::createPrior"
#> 
#> attr(,"settings")$selectorType
#> [1] "byPid"
#> 
#> attr(,"settings")$crossValidationInPrior
#> [1] TRUE
#> 
#> attr(,"settings")$modelType
#> [1] "logistic"
#> 
#> attr(,"settings")$addIntercept
#> [1] TRUE
#> 
#> attr(,"settings")$useControl
#> [1] TRUE
#> 
#> attr(,"settings")$seed
#> [1] 42
#> 
#> attr(,"settings")$name
#> [1] "Lasso Logistic Regression"
#> 
#> attr(,"settings")$threads
#> [1] -1
#> 
#> attr(,"settings")$tolerance
#> [1] 2e-06
#> 
#> attr(,"settings")$cvRepetitions
#> [1] 1
#> 
#> attr(,"settings")$maxIterations
#> [1] 3000
#> 
#> attr(,"modelType")
#> [1] "binary"
#> attr(,"saveType")
#> [1] "RtoJson"
#> 
#> attr(,"class")
#> [1] "modelSettings"
#> 
#> $splitSettings
#> $test
#> [1] 0.25
#> 
#> $train
#> [1] 0.75
#> 
#> $seed
#> [1] 42
#> 
#> $nfold
#> [1] 3
#> 
#> attr(,"fun")
#> [1] "randomSplitter"
#> attr(,"class")
#> [1] "splitSettings"
#> 
#> $executeSettings
#> $runSplitData
#> [1] TRUE
#> 
#> $runSampleData
#> [1] FALSE
#> 
#> $runFeatureEngineering
#> [1] FALSE
#> 
#> $runPreprocessData
#> [1] FALSE
#> 
#> $runModelDevelopment
#> [1] TRUE
#> 
#> $runCovariateSummary
#> [1] TRUE
#> 
#> attr(,"class")
#> [1] "executeSettings"
#> 
#> attr(,"class")
#> [1] "modelDesign"