/** Sort results to identify potential keywords. */
public void sortResults(final long max_results) {
Arrays.sort(
nodeList,
new Comparator<Node>() {
public int compare(Node n1, Node n2) {
if (n1.rank > n2.rank) {
return -1;
} else if (n1.rank < n2.rank) {
return 1;
} else {
return 0;
}
}
});
// mark the top-ranked nodes
distStats.clear();
for (int i = 0; i < nodeList.length; i++) {
final Node n1 = nodeList[i];
if (i <= max_results) {
n1.marked = true;
distStats.addValue(n1.rank);
}
if (LOG.isDebugEnabled()) {
LOG.debug("n: " + n1.key + " " + n1.rank + " " + n1.marked);
for (Node n2 : n1.edges) {
LOG.debug(" - " + n2.key);
}
}
}
}
/**
* Computes binStats
*
* @param min minimum value
* @param delta grid size
* @throws IOException if an IO error occurs
*/
@Override
public void computeBinStats(double min, double delta) throws IOException {
for (int i = 0; i < inputArray.length; i++) {
SummaryStatistics stats = binStats.get(findBin(min, inputArray[i], delta));
stats.addValue(inputArray[i]);
}
}
@Override
public double[] getResult() {
double[] result = new double[3];
result[0] = overallRes / overallQueries;
result[1] = sumStats.getStandardDeviation();
result[2] = sumStats.getVariance();
return result;
}
/**
* Calculates and returns the arrival rate of the traces in piList
*
* @return double
*/
public double getArrivalRate() {
double arrivalRate = 0;
if (arrivalStats.getN() > 0 && arrivalStats.getMean() != 0) {
// mean arrivalRate is 1 divided by the mean of the inter-arrival
// times
arrivalRate = 1 / arrivalStats.getMean();
}
return arrivalRate;
}
public static void main(String[] args) {
Ordering ordering = new Ordering();
ordering.prepare("ww3d", SequenceType.allen);
SummaryStatistics ss = new SummaryStatistics();
for (int i = 0; i < 100; ++i) {
double value = ordering.experiment(SequenceType.allen, true);
// double value = ordering.orderingExperiment("ww3d", SequenceType.allen, true);
System.out.println("..." + value);
ss.addValue(value);
}
System.out.println("Summary " + ss.getMean() + " -- " + ss.getStandardDeviation());
}
/**
* Computes binStats
*
* @param min minimum value
* @param delta grid size
* @throws IOException if an IO error occurs
*/
@Override
public void computeBinStats(double min, double delta) throws IOException {
String str = null;
double val = 0.0d;
while ((str = inputStream.readLine()) != null) {
val = Double.parseDouble(str);
SummaryStatistics stats = binStats.get(findBin(min, val, delta));
stats.addValue(val);
}
inputStream.close();
inputStream = null;
}
/**
* Computes the empirical distribution using data read from a URL.
*
* @param url url of the input file
* @throws IOException if an IO error occurs
*/
public void load(URL url) throws IOException {
BufferedReader in = new BufferedReader(new InputStreamReader(url.openStream()));
try {
DataAdapter da = new StreamDataAdapter(in);
try {
da.computeStats();
} catch (IOException ioe) {
// don't wrap exceptions which are already IOException
throw ioe;
} catch (RuntimeException rte) {
// don't wrap RuntimeExceptions
throw rte;
} catch (Exception e) {
throw MathRuntimeException.createIOException(e);
}
if (sampleStats.getN() == 0) {
throw MathRuntimeException.createEOFException("URL {0} contains no data", url);
}
in = new BufferedReader(new InputStreamReader(url.openStream()));
fillBinStats(in);
loaded = true;
} finally {
try {
in.close();
} catch (IOException ex) {
// ignore
}
}
}
@Test
public void testMemoryLeakage() {
long memoryBaseline = 0;
SummaryStatistics stats = new SummaryStatistics();
int numRuns = 25;
for (int r = 0; r < numRuns; r++) {
if (r == 1 || r == (numRuns - 1)) {
memoryBaseline = MemoryAssess.getMemoryUse();
stats.addValue(memoryBaseline);
// System.out.println("Memory before run "+(r+1)+": " + memoryBaseline / 1024 + " KB");
}
for (int t = 0; t < 1000; t++) {
graph.addVertex(null);
graph.rollback();
TitanTransaction tx = graph.newTransaction();
tx.addVertex();
tx.rollback();
}
if (r == 1 || r == (numRuns - 1)) {
memoryBaseline = MemoryAssess.getMemoryUse();
stats.addValue(memoryBaseline);
// System.out.println("Memory after run " + (r + 1) + ": " + memoryBaseline / 1024 + " KB");
}
clopen();
}
System.out.println(
"Average: " + stats.getMean() + " Std. Dev: " + stats.getStandardDeviation());
assertTrue(stats.getStandardDeviation() < stats.getMin());
}
/**
* Fills binStats array (second pass through data file).
*
* @param in object providing access to the data
* @throws IOException if an IO error occurs
*/
private void fillBinStats(Object in) throws IOException {
// Load array of bin upper bounds -- evenly spaced from min - max
double min = sampleStats.getMin();
double max = sampleStats.getMax();
double delta = (max - min) / (Double.valueOf(binCount)).doubleValue();
double[] binUpperBounds = new double[binCount];
binUpperBounds[0] = min + delta;
for (int i = 1; i < binCount - 1; i++) {
binUpperBounds[i] = binUpperBounds[i - 1] + delta;
}
binUpperBounds[binCount - 1] = max;
// Initialize binStats ArrayList
if (!binStats.isEmpty()) {
binStats.clear();
}
for (int i = 0; i < binCount; i++) {
SummaryStatistics stats = new SummaryStatistics();
binStats.add(i, stats);
}
// Filling data in binStats Array
DataAdapterFactory aFactory = new DataAdapterFactory();
DataAdapter da = aFactory.getAdapter(in);
try {
da.computeBinStats(min, delta);
} catch (IOException ioe) {
// don't wrap exceptions which are already IOException
throw ioe;
} catch (RuntimeException rte) {
// don't wrap RuntimeExceptions
throw rte;
} catch (Exception e) {
throw MathRuntimeException.createIOException(e);
}
// Assign upperBounds based on bin counts
upperBounds = new double[binCount];
upperBounds[0] = ((double) binStats.get(0).getN()) / (double) sampleStats.getN();
for (int i = 1; i < binCount - 1; i++) {
upperBounds[i] =
upperBounds[i - 1] + ((double) binStats.get(i).getN()) / (double) sampleStats.getN();
}
upperBounds[binCount - 1] = 1.0d;
}
@Override
public void finishQuery(int qryN) {
queryVal = (double) correctHits / (double) map.size();
overallRes += queryVal;
sumStats.addValue(queryVal);
map = new HashMap<Integer, Integer>();
correctHits = 0;
overallQueries++;
}
/**
* Calculates the average, min ad max throughput time out of the throughput times of all traces in
* piList. Next to this, the arrival rate is calculated. All metrics are based on the process
* instances in piList only
*
* @param piList ArrayList: the process instances used
* @param fitOption int: the fit option used (how to deal with non-conformance)
* @throws Exception
*/
public void calculateMetrics(ArrayList piList, int fitOption) throws Exception {
properFrequency = 0;
timeStats.clear();
arrivalStats.clear();
ArrayList arrivalDates = new ArrayList();
ListIterator lit = piList.listIterator();
while (lit.hasNext()) {
ExtendedLogTrace currentTrace = (ExtendedLogTrace) lit.next();
if (currentTrace.hasProperlyTerminated() && currentTrace.hasSuccessfullyExecuted()) {
properFrequency++;
}
try {
long tp = (currentTrace.getEndDate().getTime() - currentTrace.getBeginDate().getTime());
if (fitOption == 0) {
// timeStats based on all traces
timeStats.addValue(tp);
arrivalDates.add(currentTrace.getBeginDate());
}
if (currentTrace.hasProperlyTerminated() && currentTrace.hasSuccessfullyExecuted()) {
if (fitOption == 1) {
// timeStats based on fitting traces only
timeStats.addValue(tp);
arrivalDates.add(currentTrace.getBeginDate());
}
}
} catch (NullPointerException ex) {
ex.printStackTrace();
}
}
Date[] arrivals = (Date[]) arrivalDates.toArray(new Date[0]);
// make sure arrivaldates are sorted
Arrays.sort(arrivals);
if (arrivals.length > 1) {
for (int i = 1; i < arrivals.length; i++) {
long t1 = arrivals[i].getTime();
long t2 = arrivals[i - 1].getTime();
long iat = arrivals[i].getTime() - arrivals[i - 1].getTime();
if (iat >= 0) {
arrivalStats.addValue(iat);
}
}
}
}
/** Iterate through the graph, calculating rank. */
protected void iterateGraph(final int max_iterations) {
final double[] rankList = new double[nodeList.length];
// either run through N iterations, or until the standard
// error converges below a threshold
for (int k = 0; k < max_iterations; k++) {
distStats.clear();
// calculate the next rank for each node
for (int i = 0; i < nodeList.length; i++) {
final Node n1 = nodeList[i];
double rank = 0.0D;
for (Node n2 : n1.edges) {
rank += n2.rank / (double) n2.edges.size();
}
rank *= TEXTRANK_DAMPING_FACTOR;
rank += 1.0D - TEXTRANK_DAMPING_FACTOR;
rankList[i] = rank;
distStats.addValue(Math.abs(n1.rank - rank));
// System.out.println("node : " + n1.key + " rank : " + Math.abs((n1.rank - rank)));
}
final double standard_error =
distStats.getStandardDeviation() / Math.sqrt((double) distStats.getN());
// swap in new rank values
for (int i = 0; i < nodeList.length; i++) {
nodeList[i].rank = rankList[i];
}
if (standard_error < STANDARD_ERROR_THRESHOLD) {
break;
}
}
}
private GeoSegment getQQLineSegment() {
SummaryStatistics stats = new SummaryStatistics();
for (int i = 0; i < sortedData.length; i++) {
stats.addValue(sortedData[i]);
}
double sd = stats.getStandardDeviation();
double mean = stats.getMean();
double min = stats.getMin();
double max = stats.getMax();
// qq line: y = (1/sd)x - mean/sd
GeoPoint startPoint = new GeoPoint(cons);
startPoint.setCoords(min, (min / sd) - mean / sd, 1.0);
GeoPoint endPoint = new GeoPoint(cons);
endPoint.setCoords(max, (max / sd) - mean / sd, 1.0);
GeoSegment seg = new GeoSegment(cons, startPoint, endPoint);
seg.calcLength();
return seg;
}
/**
* Generates a random value from this distribution.
*
* @return the random value.
* @throws IllegalStateException if the distribution has not been loaded
*/
public double getNextValue() throws IllegalStateException {
if (!loaded) {
throw MathRuntimeException.createIllegalStateException("distribution not loaded");
}
// Start with a uniformly distributed random number in (0,1)
double x = Math.random();
// Use this to select the bin and generate a Gaussian within the bin
for (int i = 0; i < binCount; i++) {
if (x <= upperBounds[i]) {
SummaryStatistics stats = binStats.get(i);
if (stats.getN() > 0) {
if (stats.getStandardDeviation() > 0) { // more than one obs
return randomData.nextGaussian(stats.getMean(), stats.getStandardDeviation());
} else {
return stats.getMean(); // only one obs in bin
}
}
}
}
throw new MathRuntimeException("no bin selected");
}
/** * Contains all the performance results obtained during log replay analysis. Can be used to retrieve * values for the performance metrics and to get extended visualizations. * * @see PerformanceMeasurer * @author Peter T.G. Hornix ([email protected]) */ public class PerformanceLogReplayResult extends LogReplayAnalysisResult { // DescriptiveStatistics-object in which throughput times can be stored private DescriptiveStatistics timeStats = DescriptiveStatistics.newInstance(); // SummaryStatistics to obtain mean inter arrival times private SummaryStatistics arrivalStats = SummaryStatistics.newInstance(); // number of log traces that can be replayed normally private int properFrequency; public PerformanceLogReplayResult( AnalysisConfiguration analysisOptions, PetriNet net, LogReader log, LogReplayAnalysisMethod method) { // call the constructor of the superclass super(analysisOptions, net, log, method); } /** * Initializes the diagnostic data structures needed to store the measurements taken during the * log replay analysis. */ protected void initDiagnosticDataStructures() { replayedLog = new ExtendedLogReader(inputLogReader); replayedPetriNet = new ExtendedPetriNet(inputPetriNet, replayedLog.getLogTraceIDs()); } // ////////////////////////////METRICS-RELATED // METHODS/////////////////////////// /** * Calculates the average, min ad max throughput time out of the throughput times of all traces in * piList. Next to this, the arrival rate is calculated. All metrics are based on the process * instances in piList only * * @param piList ArrayList: the process instances used * @param fitOption int: the fit option used (how to deal with non-conformance) * @throws Exception */ public void calculateMetrics(ArrayList piList, int fitOption) throws Exception { properFrequency = 0; timeStats.clear(); arrivalStats.clear(); ArrayList arrivalDates = new ArrayList(); ListIterator lit = piList.listIterator(); while (lit.hasNext()) { ExtendedLogTrace currentTrace = (ExtendedLogTrace) lit.next(); if (currentTrace.hasProperlyTerminated() && currentTrace.hasSuccessfullyExecuted()) { properFrequency++; } try { long tp = (currentTrace.getEndDate().getTime() - currentTrace.getBeginDate().getTime()); if (fitOption == 0) { // timeStats based on all traces timeStats.addValue(tp); arrivalDates.add(currentTrace.getBeginDate()); } if (currentTrace.hasProperlyTerminated() && currentTrace.hasSuccessfullyExecuted()) { if (fitOption == 1) { // timeStats based on fitting traces only timeStats.addValue(tp); arrivalDates.add(currentTrace.getBeginDate()); } } } catch (NullPointerException ex) { ex.printStackTrace(); } } Date[] arrivals = (Date[]) arrivalDates.toArray(new Date[0]); // make sure arrivaldates are sorted Arrays.sort(arrivals); if (arrivals.length > 1) { for (int i = 1; i < arrivals.length; i++) { long t1 = arrivals[i].getTime(); long t2 = arrivals[i - 1].getTime(); long iat = arrivals[i].getTime() - arrivals[i - 1].getTime(); if (iat >= 0) { arrivalStats.addValue(iat); } } } } /** * Exports the throughput times of all process instances in piList to a comma-seperated text-file. * * @param piList ArrayList: the process instances used * @param file File: the file to which the times are exported * @param divider long: the time divider used * @param sort String: the time sort used * @param fitOption int: the fit option used (how to deal with non-conformance) * @throws IOException */ public void exportToFile(ArrayList piList, File file, long divider, String sort, int fitOption) throws IOException { Writer output = new BufferedWriter(new FileWriter(file)); String line = "Log Trace,Throughput time (" + sort + ")\n"; output.write(line); ListIterator lit = piList.listIterator(); while (lit.hasNext()) { ExtendedLogTrace currentTrace = (ExtendedLogTrace) lit.next(); try { double tp = (currentTrace.getEndDate().getTime() - currentTrace.getBeginDate().getTime()) * 1.0 / divider; if (fitOption == 0) { // times based on all traces line = currentTrace.getName() + "," + tp + "\n"; // write line to the file output.write(line); } if (fitOption == 1 && currentTrace.hasProperlyTerminated() && currentTrace.hasSuccessfullyExecuted()) { // times based on fitting traces only line = currentTrace.getName() + "," + tp + "\n"; // write line to the file output.write(line); } } catch (NullPointerException npe) { } } // close the file output.close(); } // ////////////////////////////GET // METHODS/////////////////////////////////////// /** * Calculates and returns the stdev in throughput time out of the throughput times in timeStats. * (make sure calculateProcessMetrics() is called before this method). * * @return double */ public double getStdevThroughputTime() { return timeStats.getStandardDeviation(); } /** * Calculates the average of the (fastestpercentage) fast traces, the (slowestPercentage) slow * traces and the (100% - fastestPercentage - slowestPercentage) normal speed traces and returns * these averages in an array, where [0]: avg fast throughput time [1]: avg slow throughput time * [2]: avg middle throughput time * * @param fastestPercentage double: the percentage of measurements that is to be counted as fast * @param slowestPercentage double: the percentage of measurements that is to be counted as slow * @return double[] */ public double[] getAverageTimes(double fastestPercentage, double slowestPercentage) { // initialize arrays double[] timeList = timeStats.getSortedValues(); double[] avgTimes = new double[3]; long total = 0; // obtain the number of fast , slow, normal traces int[] sizes = getSizes(fastestPercentage, slowestPercentage); int fastSize = sizes[0], slowSize = sizes[1], middleSize = sizes[2]; for (int i = 0; i < fastSize; i++) { total += timeList[i]; } // calculate average of the fastest traces double avgFastestTime = 0.0; if (fastSize != 0) { avgFastestTime = (total * 1.0) / fastSize; } // calculate average of the slowest traces int upperSize = timeList.length - slowSize; total = 0; for (int i = upperSize; i < timeList.length; i++) { total += timeList[i]; } double avgSlowestTime = 0.0; if (slowSize > 0) { avgSlowestTime = (total * 1.0) / slowSize; } // calculate the middle/normal-speed traces total = 0; for (int i = fastSize; i < upperSize; i++) { total += timeList[i]; } double avgMiddleTime = 0.0; if (middleSize > 0) { avgMiddleTime = (total * 1.0) / middleSize; } avgTimes[0] = avgFastestTime; avgTimes[1] = avgSlowestTime; avgTimes[2] = avgMiddleTime; return avgTimes; } /** * Returns an array containing the number of process instances that are considered to be fast, * i.e. have a low throughput time (place 0 in array), the number of process instances that are * slow (place 1 in array) and the number of process instances that are considered to be of normal * speed (place 2 in array). Based on fastestPercentage, slowestPercentage and timeList (thus * method calculateProcessMetrics() should be called before this one) * * @param fastestPercentage double: the percentage of measurements that is to be counted as fast * @param slowestPercentage double: the percentage of measurements that is to be counted as slow * @return int[] */ public int[] getSizes(double fastestPercentage, double slowestPercentage) { int[] sizes = new int[3]; String sizeString; int length = timeStats.getValues().length; sizeString = Math.round((length * fastestPercentage) / 100.0) + ""; sizes[0] = Integer.parseInt(sizeString); if (sizes[0] != length) { sizeString = Math.round((length * slowestPercentage) / 100.0) + ""; sizes[1] = Integer.parseInt(sizeString); if ((sizes[0] + sizes[1]) > length) { // Make sure that sizes[0] + sizes[1] remains smaller than // the number of measurements in timeList (rounding could mess // this up) sizes[1] = length - sizes[0]; } } else { sizes[1] = 0; } sizes[2] = length - sizes[0] - sizes[1]; return sizes; } /** * Calculates and returns the arrival rate of the traces in piList * * @return double */ public double getArrivalRate() { double arrivalRate = 0; if (arrivalStats.getN() > 0 && arrivalStats.getMean() != 0) { // mean arrivalRate is 1 divided by the mean of the inter-arrival // times arrivalRate = 1 / arrivalStats.getMean(); } return arrivalRate; } /** * Returns the arrival Stats of the traces in piList * * @return SummaryStatistics */ public SummaryStatistics getArrivalStats() { return arrivalStats; } /** * Returns the mean throughput time * * @return double */ public double getMeanThroughputTime() { return timeStats.getMean(); } /** * Returns the minimal throughput time. Note that method calculateProcessMetrics() should be * called before this method. * * @return double */ public double getMinThroughputTime() { return timeStats.getMin(); } /** * Returns the maximal throughput time * * @return double */ public double getMaxThroughputTime() { return timeStats.getMax(); } /** * returns the number of cases that execute successfully and complete properly * * @return int */ public int getProperFrequency() { return (properFrequency); } // ////////////////////////////GRAPPA-RELATED // METHODS/////////////////////////// /** * Creates a visualization of the performance analysis results. Note that a change of the display * state by the user will have no effect before calling this methods. This is intended to prevent * unnecessary cloning of the extended petri net, which actually delivers the custom visualization * of the performance analysis results. * * @param selectedInstances The process instances that have been selected for updating the * visualization. * @return The visualization wrapped in a ModelGraphPanel. */ public ModelGraphPanel getVisualization(ArrayList selectedInstances) { // sets the currentlySelectedInstances attribute, which is necessary // because // the writeToDot() method has a fixed interface, though the // visualization should // be able to take them into account ((ExtendedPetriNet) replayedPetriNet).currentlySelectedInstances = selectedInstances; ModelGraphPanel myResultVisualization; myResultVisualization = ((ExtendedPetriNet) replayedPetriNet).getGrappaVisualization(); return myResultVisualization; } }
/** Calculate a threshold for the ranked results. */
public double getRankThreshold() {
return distStats.getMean() + (distStats.getStandardDeviation() * INCLUSIVE_COEFF);
}
/**
* Calculates the rank weighting scores for all the nodes in the graph. Iterative calculates over
* the graph until convergence at the standard error threshold or until max iterations.
*
* @param maxIterations Max number of iterations allowed for calculating rank scores
* @param language Language of the text to calculate rank weighting scores for Available
* languages: Icelandic and English
*/
public void weigthingScore(int maxIterations, Language language) {
LinkedList<Node> nodes = new LinkedList<Node>();
// Add nodes to LinkedList, we need them to stay in order
for (int i = 0; i < nodeList.length; i++) {
nodes.add(nodeList[i]);
}
/*
WS(Vi) = ( 1 - d) + d * Sum(VjIn) ________Wij________ * WS(Vj)
Sum(Vk outVj) Wjk
*/
for (int k = 0; k < maxIterations; k++) {
distStats.clear();
// Use dynamic programming to calculate the scores
double previousWSScore[] = new double[nodes.size()];
// Read in scores already calculated for nodes
for (Node s : nodeList) {
previousWSScore[nodes.indexOf(s)] = s.rank;
}
// For all nodes in the graph
for (Node sentence_i : nodes) {
double resultSumVji = 0;
// For all in-coming edges of Vi
for (Node sentence_j : sentence_i.edgesIN) {
// Do not compare a sentence to it self, we do not allow self voting here
if (!sentence_j.value.text.equalsIgnoreCase(sentence_i.value.text)) {
// Calculate the sum of all similarity measurements
// from all Vj nodes with outgoing edges to Vk nodes, see Wjk in equation
double sumWjk = getSumWjk(sentence_j, language);
if (sumWjk != 0) {
double Wji = 0.0;
if (language.equals(Language.ICELANDIC)) {
// Calculate Wij, similarity between two sentences
Wji = sentence_i.similarity(sentence_j);
} else if (language.equals(Language.ENGLISH)) {
// Calculate Wij, similarity between two sentences
Wji = sentence_i.similarityEN(sentence_j);
}
// Get the score for the previous node
double WSVj = previousWSScore[nodes.indexOf(sentence_j)];
// Sum all (j in Vj)
resultSumVji += ((Wji / sumWjk) * WSVj);
}
}
}
// Calculate weighting score WS(Vi)
double WSVi = (1.0 - TEXTRANK_DAMPING_FACTOR) + TEXTRANK_DAMPING_FACTOR * resultSumVji;
distStats.addValue(Math.abs(sentence_i.rank - WSVi));
sentence_i.rank = WSVi;
}
// Calculate the Standard Error of the Mean
final double standard_error =
distStats.getStandardDeviation() / Math.sqrt((double) distStats.getN());
// if std error of the mean is less than threshold
// the graph has converged and we break
if (standard_error < STANDARD_ERROR_THRESHOLD) {
break;
}
}
}