public void Regression_linear_multivariable_web_example_single_row_0()
{
InitData_dataset_regression_web_example();
BuildLinearMultiVariable lm = new BuildLinearMultiVariable();
ModelLinearMultiVariableBase mb = (ModelLinearMultiVariableBase)lm.BuildModel(_trainingData,
_attributeHeaders, _indexTargetAttribute);
double[] data = GetSingleTrainingRowDataForTest(0);
double value = mb.RunModelForSingleData(data);
Assert.IsTrue(SupportFunctions.DoubleCompare(value, 249.98));
}
public void Regression_linear_multivariable_jason_simple()
{
Init_dataset_jason_linear_regression();
BuildLinearMultiVariable lm = new BuildLinearMultiVariable();
ModelLinearMultiVariableBase mb = (ModelLinearMultiVariableBase)lm.BuildModel(_trainingData,
_attributeHeaders, _indexTargetAttribute);
double[] data = { 1 };
double value = mb.RunModelForSingleData(data);
Assert.IsTrue(SupportFunctions.DoubleCompare(value, 1.2));
}
public void Regression_linear_multivariable_jason_simple_rmse()
{
Init_dataset_jason_linear_regression();
BuildLinearMultiVariable lm = new BuildLinearMultiVariable();
ModelLinearMultiVariableBase mb = (ModelLinearMultiVariableBase)lm.BuildModel(_trainingData,
_attributeHeaders, _indexTargetAttribute);
//Refill the trainingData array since it gets messed up
Init_dataset_jason_linear_regression();
double value = mb.GetModelRMSE(_trainingData);
Assert.IsTrue(SupportFunctions.DoubleCompare(value,
0.69));
}
public void Regression_linear_multivariable_pythagoras_row_5()
{
Init_dataset_pythagoras();
BuildLinearMultiVariable lm = new BuildLinearMultiVariable();
ModelLinearMultiVariableBase mb = (ModelLinearMultiVariableBase)
lm.BuildModel(
_trainingData,
_attributeHeaders,
_indexTargetAttribute);
int row = 5;
double[] data = GetSingleTrainingRowDataForTest(row);
double value = mb.RunModelForSingleData(data);
//Value is within +-1.0
Assert.IsTrue(value <
_trainingData[_indexTargetAttribute][row] + 1.0 &&
value> _trainingData[_indexTargetAttribute][row] - 1.0);
}
