RFTrainTestTurbineModel: Train and test a random forest power model

In AndyClifton/PowerPerformance: Package to analyze wind resource and wind turbine data (developer test version)

Description

RFTrainTestTurbineModel trains and then tests a random forest power model of a wind turbine using actual observations. The function returns basic error and CPU usage metrics. This function is intended for assessing the accuracy of a particular model.

Usage

RFTrainTestTurbineModel(train.predictors, train.response, test.predictors,
  test.response)

Arguments

train.predictors	a data.frame containing the predictors that will be used to train the model
train.response	a vector containing the response of the turbine (this could be power, loads, ...)
test.predictors	a data.frame containing a new set of predictors that will be use to create a new response matrix
test.response	a vector of observed responses

Value

a list containing

model

the random forest model

metrics

a data.frame containing different error and performance metrics

See Also

GetErrorMetrics

Other Random forest methods: RFCheckVarImp; RFChecknTrees; RFQueryTurbineModel; RFTrainTurbineModel

AndyClifton/PowerPerformance documentation built on May 5, 2019, 6 a.m.