public void RegressionForestModel_PredictCertainty_Single()
{
var parser = new CsvParser(() => new StringReader(Resources.AptitudeData));
var observations = parser.EnumerateRows(v => v != "Pass").ToF64Matrix();
var targets = parser.EnumerateRows("Pass").ToF64Vector();
var rows = targets.Length;
var learner = new RegressionRandomForestLearner(100, 5, 100, 1, 0.0001, 1.0, 42, false);
var sut = learner.Learn(observations, targets);
var actual = new CertaintyPrediction[rows];
for (int i = 0; i < rows; i++)
{
actual[i] = sut.PredictCertainty(observations.Row(i));
}
var evaluator = new MeanSquaredErrorRegressionMetric();
var error = evaluator.Error(targets, actual.Select(p => p.Prediction).ToArray());
Assert.AreEqual(0.15381141277554411, error, 0.0000001);
var expected = new CertaintyPrediction[] { new CertaintyPrediction(0.379151515151515, 0.0608255007215007), new CertaintyPrediction(0.411071351850763, 0.0831655436577049), new CertaintyPrediction(0.243420918950331, 0.0452827034233046), new CertaintyPrediction(0.302332251082251, 0.0699917594408057), new CertaintyPrediction(0.411071351850763, 0.0831655436577049), new CertaintyPrediction(0.175743762773174, 0.0354069437824887), new CertaintyPrediction(0.574083361083361, 0.0765858693929188), new CertaintyPrediction(0.259063776093188, 0.0491198812971218), new CertaintyPrediction(0.163878898878899, 0.0331543420321184), new CertaintyPrediction(0.671753996003996, 0.0624466591504497), new CertaintyPrediction(0.418472943722944, 0.0607014359023913), new CertaintyPrediction(0.243420918950331, 0.0452827034233046), new CertaintyPrediction(0.443779942279942, 0.0941961872991865), new CertaintyPrediction(0.156999361749362, 0.0435804333960299), new CertaintyPrediction(0.591222034501446, 0.0873624628347336), new CertaintyPrediction(0.123822406351818, 0.0283119805431255), new CertaintyPrediction(0.162873993653405, 0.0333697457759022), new CertaintyPrediction(0.596261932511932, 0.0695341060210394), new CertaintyPrediction(0.671753996003996, 0.0624466591504497), new CertaintyPrediction(0.418472943722944, 0.0607014359023913), new CertaintyPrediction(0.329000027750028, 0.0788869852405852), new CertaintyPrediction(0.671753996003996, 0.0624466591504497), new CertaintyPrediction(0.499770375049787, 0.0913884936411888), new CertaintyPrediction(0.140025508804921, 0.0309875116490099), new CertaintyPrediction(0.161207326986739, 0.0336321035325246), new CertaintyPrediction(0.389553418803419, 0.0744433596104835), };
CollectionAssert.AreEqual(expected, actual);
}
/// <summary>
/// Predicts a set of obervations with certainty predictions
/// </summary>
/// <param name="observations"></param>
/// <returns></returns>
public CertaintyPrediction[] PredictCertainty(F64Matrix observations)
{
var rows = observations.RowCount;
var predictions = new CertaintyPrediction[rows];
for (int i = 0; i < rows; i++)
{
predictions[i] = PredictCertainty(observations.Row(i));
}
return(predictions);
}
