/// <summary>
/// Runs the module.
/// </summary>
/// <param name="args">The command line arguments for the module.</param>
/// <param name="usagePrefix">The prefix to print before the usage string.</param>
/// <returns>True if the run was successful, false otherwise.</returns>
public override bool Run(string[] args, string usagePrefix)
{
string groundTruthFileName = string.Empty;
string predictionsFileName = string.Empty;
string reportFileName = string.Empty;
string calibrationCurveFileName = string.Empty;
string rocCurveFileName = string.Empty;
string precisionRecallCurveFileName = string.Empty;
string positiveClassLabel = string.Empty;
var parser = new CommandLineParser();
parser.RegisterParameterHandler("--ground-truth", "FILE", "File with ground truth labels", v => groundTruthFileName = v, CommandLineParameterType.Required);
parser.RegisterParameterHandler("--predictions", "FILE", "File with label predictions", v => predictionsFileName = v, CommandLineParameterType.Required);
parser.RegisterParameterHandler("--report", "FILE", "File to store the evaluation report", v => reportFileName = v, CommandLineParameterType.Optional);
parser.RegisterParameterHandler("--calibration-curve", "FILE", "File to store the empirical calibration curve", v => calibrationCurveFileName = v, CommandLineParameterType.Optional);
parser.RegisterParameterHandler("--roc-curve", "FILE", "File to store the receiver operating characteristic curve", v => rocCurveFileName = v, CommandLineParameterType.Optional);
parser.RegisterParameterHandler("--precision-recall-curve", "FILE", "File to store the precision-recall curve", v => precisionRecallCurveFileName = v, CommandLineParameterType.Optional);
parser.RegisterParameterHandler("--positive-class", "STRING", "Label of the positive class to use in curves", v => positiveClassLabel = v, CommandLineParameterType.Optional);
if (!parser.TryParse(args, usagePrefix))
{
return(false);
}
// Read ground truth
var groundTruth = ClassifierPersistenceUtils.LoadLabeledFeatureValues(groundTruthFileName);
// Read predictions using ground truth label dictionary
var predictions = ClassifierPersistenceUtils.LoadLabelDistributions(predictionsFileName, groundTruth.First().LabelDistribution.LabelSet);
// Check that there are at least two distinct class labels
if (predictions.First().LabelSet.Count < 2)
{
throw new InvalidFileFormatException("Ground truth and predictions must contain at least two distinct class labels.");
}
// Distill distributions and point estimates
var predictiveDistributions = predictions.Select(i => i.ToDictionary()).ToList();
var predictivePointEstimates = predictions.Select(i => i.GetMode()).ToList();
// Create evaluator
var evaluatorMapping = Mappings.Classifier.ForEvaluation();
var evaluator = new ClassifierEvaluator <IList <LabeledFeatureValues>, LabeledFeatureValues, IList <LabelDistribution>, string>(evaluatorMapping);
// Write evaluation report
if (!string.IsNullOrEmpty(reportFileName))
{
using (var writer = new StreamWriter(reportFileName))
{
this.WriteReportHeader(writer, groundTruthFileName, predictionsFileName);
this.WriteReport(writer, evaluator, groundTruth, predictiveDistributions, predictivePointEstimates);
}
}
// Compute and write the empirical probability calibration curve
positiveClassLabel = this.CheckPositiveClassLabel(groundTruth, positiveClassLabel);
if (!string.IsNullOrEmpty(calibrationCurveFileName))
{
this.WriteCalibrationCurve(calibrationCurveFileName, evaluator, groundTruth, predictiveDistributions, positiveClassLabel);
}
// Compute and write the precision-recall curve
if (!string.IsNullOrEmpty(precisionRecallCurveFileName))
{
this.WritePrecisionRecallCurve(precisionRecallCurveFileName, evaluator, groundTruth, predictiveDistributions, positiveClassLabel);
}
// Compute and write the receiver operating characteristic curve
if (!string.IsNullOrEmpty(rocCurveFileName))
{
this.WriteRocCurve(rocCurveFileName, evaluator, groundTruth, predictiveDistributions, positiveClassLabel);
}
return(true);
}
/// <summary>
/// Runs the module.
/// </summary>
/// <param name="args">The command line arguments for the module.</param>
/// <param name="usagePrefix">The prefix to print before the usage string.</param>
/// <returns>True if the run was successful, false otherwise.</returns>
public override bool Run(string[] args, string usagePrefix)
{
string dataSetFile = string.Empty;
string resultsFile = string.Empty;
int crossValidationFoldCount = 5;
int iterationCount = BayesPointMachineClassifierTrainingSettings.IterationCountDefault;
int batchCount = BayesPointMachineClassifierTrainingSettings.BatchCountDefault;
bool computeModelEvidence = BayesPointMachineClassifierTrainingSettings.ComputeModelEvidenceDefault;
var parser = new CommandLineParser();
parser.RegisterParameterHandler("--data-set", "FILE", "File with training data", v => dataSetFile = v, CommandLineParameterType.Required);
parser.RegisterParameterHandler("--results", "FILE", "File with cross-validation results", v => resultsFile = v, CommandLineParameterType.Required);
parser.RegisterParameterHandler("--folds", "NUM", "Number of cross-validation folds (defaults to " + crossValidationFoldCount + ")", v => crossValidationFoldCount = v, CommandLineParameterType.Optional);
parser.RegisterParameterHandler("--iterations", "NUM", "Number of training algorithm iterations (defaults to " + iterationCount + ")", v => iterationCount = v, CommandLineParameterType.Optional);
parser.RegisterParameterHandler("--batches", "NUM", "Number of batches to split the training data into (defaults to " + batchCount + ")", v => batchCount = v, CommandLineParameterType.Optional);
parser.RegisterParameterHandler("--compute-evidence", "Compute model evidence (defaults to " + computeModelEvidence + ")", () => computeModelEvidence = true);
if (!parser.TryParse(args, usagePrefix))
{
return(false);
}
// Load and shuffle data
var dataSet = ClassifierPersistenceUtils.LoadLabeledFeatureValues(dataSetFile);
BayesPointMachineClassifierModuleUtilities.WriteDataSetInfo(dataSet);
Rand.Restart(562);
Rand.Shuffle(dataSet);
// Create evaluator
var evaluatorMapping = Mappings.Classifier.ForEvaluation();
var evaluator = new ClassifierEvaluator <IList <LabeledFeatureValues>, LabeledFeatureValues, IList <LabelDistribution>, string>(evaluatorMapping);
// Create performance metrics
var accuracy = new List <double>();
var negativeLogProbability = new List <double>();
var auc = new List <double>();
var evidence = new List <double>();
var iterationCounts = new List <double>();
var trainingTime = new List <double>();
// Run cross-validation
int validationSetSize = dataSet.Count / crossValidationFoldCount;
Console.WriteLine("Running {0}-fold cross-validation on {1}", crossValidationFoldCount, dataSetFile);
// TODO: Use chained mapping to implement cross-validation
for (int fold = 0; fold < crossValidationFoldCount; fold++)
{
// Construct training and validation sets for fold
int validationSetStart = fold * validationSetSize;
int validationSetEnd = (fold + 1 == crossValidationFoldCount)
? dataSet.Count
: (fold + 1) * validationSetSize;
var trainingSet = new List <LabeledFeatureValues>();
var validationSet = new List <LabeledFeatureValues>();
for (int instance = 0; instance < dataSet.Count; instance++)
{
if (validationSetStart <= instance && instance < validationSetEnd)
{
validationSet.Add(dataSet[instance]);
}
else
{
trainingSet.Add(dataSet[instance]);
}
}
// Print info
Console.WriteLine(" Fold {0} [validation set instances {1} - {2}]", fold + 1, validationSetStart, validationSetEnd - 1);
// Create classifier
var classifier = BayesPointMachineClassifier.CreateBinaryClassifier(Mappings.Classifier);
classifier.Settings.Training.IterationCount = iterationCount;
classifier.Settings.Training.BatchCount = batchCount;
classifier.Settings.Training.ComputeModelEvidence = computeModelEvidence;
int currentIterationCount = 0;
classifier.IterationChanged += (sender, eventArgs) => { currentIterationCount = eventArgs.CompletedIterationCount; };
// Train classifier
var stopWatch = new Stopwatch();
stopWatch.Start();
classifier.Train(trainingSet);
stopWatch.Stop();
// Produce predictions
var predictions = classifier.PredictDistribution(validationSet).ToList();
var predictedLabels = predictions.Select(
prediction => prediction.Aggregate((aggregate, next) => next.Value > aggregate.Value ? next : aggregate).Key).ToList();
// Iteration count
iterationCounts.Add(currentIterationCount);
// Training time
trainingTime.Add(stopWatch.ElapsedMilliseconds);
// Compute accuracy
accuracy.Add(1 - (evaluator.Evaluate(validationSet, predictedLabels, Metrics.ZeroOneError) / predictions.Count));
// Compute mean negative log probability
negativeLogProbability.Add(evaluator.Evaluate(validationSet, predictions, Metrics.NegativeLogProbability) / predictions.Count);
// Compute M-measure (averaged pairwise AUC)
auc.Add(evaluator.AreaUnderRocCurve(validationSet, predictions));
// Compute log evidence if desired
evidence.Add(computeModelEvidence ? classifier.LogModelEvidence : double.NaN);
// Persist performance metrics
Console.WriteLine(
" Accuracy = {0,5:0.0000} NegLogProb = {1,5:0.0000} AUC = {2,5:0.0000}{3} Iterations = {4} Training time = {5}",
accuracy[fold],
negativeLogProbability[fold],
auc[fold],
computeModelEvidence ? string.Format(" Log evidence = {0,5:0.0000}", evidence[fold]) : string.Empty,
iterationCounts[fold],
BayesPointMachineClassifierModuleUtilities.FormatElapsedTime(trainingTime[fold]));
BayesPointMachineClassifierModuleUtilities.SavePerformanceMetrics(
resultsFile, accuracy, negativeLogProbability, auc, evidence, iterationCounts, trainingTime);
}
return(true);
}