public double[][] trainWeights(
GeneralDataset<L, F> dataset,
double[] initial,
boolean bypassTuneSigma,
Minimizer<DiffFunction> minimizer) {
if (minimizer == null) minimizer = minimizerCreator.create();
if (dataset instanceof RVFDataset) ((RVFDataset<L, F>) dataset).ensureRealValues();
double[] interimWeights = null;
if (!bypassTuneSigma) {
if (tuneSigmaHeldOut) {
interimWeights =
heldOutSetSigma(
dataset); // the optimum interim weights from held-out training data have already
// been found.
} else if (tuneSigmaCV) {
crossValidateSetSigma(
dataset, folds); // TODO: assign optimum interim weights as part of this process.
}
}
LogConditionalObjectiveFunction<L, F> objective =
new LogConditionalObjectiveFunction<L, F>(dataset, logPrior);
if (initial == null && interimWeights != null && !retrainFromScratchAfterSigmaTuning) {
// System.err.println("## taking advantage of interim weights as starting point.");
initial = interimWeights;
}
if (initial == null) {
initial = objective.initial();
}
double[] weights = minimizer.minimize(objective, TOL, initial);
return objective.to2D(weights);
}
public void crossValidateSetSigma(
GeneralDataset<L, F> dataset, int kfold, LineSearcher minimizer) {
crossValidateSetSigma(
dataset,
kfold,
new MultiClassAccuracyStats<L>(MultiClassAccuracyStats.USE_LOGLIKELIHOOD),
minimizer);
}
/**
* callls the method {@link #crossValidateSetSigma(GeneralDataset, int, Scorer, LineSearcher)}
* with multi-class log-likelihood scoring (see {@link MultiClassAccuracyStats}) and
* golden-section line search (see {@link GoldenSectionLineSearch}).
*
* @param dataset the data set to optimize sigma on.
*/
public void crossValidateSetSigma(GeneralDataset<L, F> dataset, int kfold) {
System.err.println("##you are here.");
crossValidateSetSigma(
dataset,
kfold,
new MultiClassAccuracyStats<L>(MultiClassAccuracyStats.USE_LOGLIKELIHOOD),
new GoldenSectionLineSearch(true, 1e-2, min, max));
}
public void crossValidateSetSigma(
GeneralDataset<L, F> dataset, int kfold, final Scorer<L> scorer) {
crossValidateSetSigma(
dataset, kfold, scorer, new GoldenSectionLineSearch(true, 1e-2, min, max));
}
/**
* Calls the method {@link #crossValidateSetSigma(GeneralDataset, int)} with 5-fold
* cross-validation.
*
* @param dataset the data set to optimize sigma on.
*/
public void crossValidateSetSigma(GeneralDataset<L, F> dataset) {
crossValidateSetSigma(dataset, 5);
}