List of usage examples for org.apache.mahout.classifier.evaluation Auc add
public void add(int trueValue, int predictedClass)
From source file:br.com.sitedoph.mahout_examples.BankMarketingClassificationMain.java
License:Apache License
private static double evaluateTheCallAndGetBiggestScore(double biggestScore, OnlineLogisticRegression lr, Auc eval, TelephoneCall call) { final double score = lr.classifyScalar(call.asVector()); eval.add(call.getTarget(), score); if (score > biggestScore) { System.out.println("### SCORE > BIGGESTSCORE ### score: " + score + " accuracy " + eval.auc() + " call fields: " + call.getFields()); biggestScore = score;/* w w w . java 2s .c om*/ } return biggestScore; }
From source file:com.ml.ira.algos.RunLogistic.java
License:Apache License
static void mainToOutput(String[] args, PrintWriter output) throws Exception { if (parseArgs(args)) { if (!showAuc && !showConfusion && !showScores) { showAuc = true;/*from ww w . j a v a2s . c o m*/ showConfusion = true; } Auc collector = new Auc(); LogisticModelParameters lmp; if (modelFile.startsWith("hdfs://")) { lmp = LogisticModelParameters.loadFrom(new Path(modelFile)); } else { lmp = LogisticModelParameters.loadFrom(new File(modelFile)); } CsvRecordFactory csv = lmp.getCsvRecordFactory(); OnlineLogisticRegression lr = lmp.createRegression(); BufferedReader in = TrainLogistic.open(inputFile); //String line = in.readLine(); //csv.firstLine(line); String line; if (fieldNames != null && fieldNames.equalsIgnoreCase("internal")) { csv.firstLine(lmp.getFieldNames()); } else { csv.firstLine(in.readLine()); } line = in.readLine(); if (showScores) { output.println("\"target\",\"model-output\",\"log-likelihood\""); } while (line != null) { Vector v = new SequentialAccessSparseVector(lmp.getNumFeatures()); int target = csv.processLine(line, v); double score = lr.classifyScalar(v); if (showScores) { output.printf(Locale.ENGLISH, "%d,%.3f,%.6f%n", target, score, lr.logLikelihood(target, v)); } collector.add(target, score); line = in.readLine(); } if (showAuc) { output.printf(Locale.ENGLISH, "AUC = %.2f%n", collector.auc()); } if (showConfusion) { Matrix m = collector.confusion(); output.printf(Locale.ENGLISH, "confusion: [[%.1f, %.1f], [%.1f, %.1f]]%n", m.get(0, 0), m.get(1, 0), m.get(0, 1), m.get(1, 1)); m = collector.entropy(); output.printf(Locale.ENGLISH, "entropy: [[%.1f, %.1f], [%.1f, %.1f]]%n", m.get(0, 0), m.get(1, 0), m.get(0, 1), m.get(1, 1)); } } }
From source file:guipart.view.GUIOverviewController.java
@FXML void handleClassifyModel(ActionEvent event) throws IOException { if (pathModel != null && pathCSV != null) { Auc collector = new Auc(); LogisticModelParameters lmp = LogisticModelParameters.loadFrom(new File(pathModel)); CsvRecordFactory csv = lmp.getCsvRecordFactory(); OnlineLogisticRegression lr = lmp.createRegression(); BufferedReader in = Utils.open(pathCSV); String line = in.readLine(); csv.firstLine(line);//from w ww . ja v a 2 s. com line = in.readLine(); int correct = 0; int wrong = 0; Boolean booltemp; String gender; while (line != null) { Vector v = new SequentialAccessSparseVector(lmp.getNumFeatures()); int target = csv.processLine(line, v); String[] split = line.split(","); double score = lr.classifyFull(v).maxValueIndex(); if (score == target) correct++; else wrong++; System.out.println("Target is: " + target + " Score: " + score); booltemp = score != 0; if (split[1].contentEquals("1")) gender = "male"; else gender = "female"; Person temp = new Person(Integer.parseInt(split[0]), Integer.parseInt(split[4]), Integer.parseInt(split[7]), booltemp, gender, Integer.parseInt(split[5]), Integer.parseInt(split[6]), Integer.parseInt(split[3])); guiPart.addPerson(temp); line = in.readLine(); collector.add(target, score); } double posto = ((double) wrong / (double) (correct + wrong)) * 100; System.out.println("Total: " + (correct + wrong) + " Correct: " + correct + " Wrong: " + wrong + " Wrong pct: " + posto + "%"); //PrintWriter output = null; Matrix m = collector.confusion(); //output.printf(Locale.ENGLISH, "confusion: [[%.1f, %.1f], [%.1f, %.1f]]%n",m.get(0, 0), m.get(1, 0), m.get(0, 1), m.get(1, 1)); System.out.println("Confusion:" + m.get(0, 0) + " " + m.get(1, 0) + "\n \t " + m.get(0, 1) + " " + m.get(1, 1) + " "); // m = collector.entropy(); //output.printf(Locale.ENGLISH, "entropy: [[%.1f, %.1f], [%.1f, %.1f]]%n",m.get(0, 0), m.get(1, 0), m.get(0, 1), m.get(1, 1)); textAnalyze2.setText("Confusion:" + m.get(0, 0) + " " + m.get(1, 0) + "\n \t \t " + m.get(0, 1) + " " + m.get(1, 1) + "\n" + "Total: " + (correct + wrong) + " Correct: " + correct + " Wrong: " + wrong + " Wrong pct: " + posto + "%"); } else { Dialogs.create().owner(guiPart.getPrimaryStage()).title("Error Dialog") .masthead("Look, an Error Dialog").message("One or more files aren't selected").showError(); } }
From source file:haflow.component.mahout.logistic.RunLogistic.java
License:Apache License
static void mainToOutput(String[] args) throws Exception { if (parseArgs(args)) { if (!showAuc && !showConfusion && !showScores) { showAuc = true;//from ww w . j a va 2 s. c o m showConfusion = true; } //PrintWriter output=new PrintWriter(new FileOutputStream(outputFile),true); PrintWriter output = new PrintWriter(HdfsUtil.writeHdfs(outputFile), true); PrintWriter acc_output = new PrintWriter(HdfsUtil.writeHdfs(accurateFile), true); Auc collector = new Auc(); LogisticModelParameters lmp = LogisticModelParameters.loadFrom(HdfsUtil.open(modelFile)); CsvRecordFactory csv = lmp.getCsvRecordFactory(); OnlineLogisticRegression lr = lmp.createRegression(); BufferedReader in = new BufferedReader(new InputStreamReader(HdfsUtil.open(inputFile))); String line = in.readLine(); csv.firstLine(line); line = in.readLine(); if (showScores) { output.println("\"target\",\"model-output\",\"log-likelihood\""); } while (line != null) { Vector v = new SequentialAccessSparseVector(lmp.getNumFeatures()); int target = csv.processLine(line, v); double score = lr.classifyScalar(v); if (showScores) { output.printf(Locale.ENGLISH, "%d,%.3f,%.6f%n", target, score, lr.logLikelihood(target, v)); } collector.add(target, score); line = in.readLine(); } if (showAuc) { acc_output.printf(Locale.ENGLISH, "AUC , %.2f%n", collector.auc()); } if (showConfusion) { Matrix m = collector.confusion(); acc_output.printf(Locale.ENGLISH, "confusion, [[%.1f %.1f], [%.1f %.1f]]%n", m.get(0, 0), m.get(1, 0), m.get(0, 1), m.get(1, 1)); m = collector.entropy(); acc_output.printf(Locale.ENGLISH, "entropy, [[%.1f %.1f], [%.1f %.1f]]%n", m.get(0, 0), m.get(1, 0), m.get(0, 1), m.get(1, 1)); } output.close(); acc_output.close(); } }
From source file:javaapplication3.RunLogistic.java
public static void main(String[] args) throws IOException { // TODO code application logic here Auc collector = new Auc(); LogisticModelParameters lmp = LogisticModelParameters.loadFrom(new File(modelFile)); CsvRecordFactory csv = lmp.getCsvRecordFactory(); OnlineLogisticRegression lr = lmp.createRegression(); BufferedReader in = open(inputFile); String line = in.readLine();//ww w . j a v a 2 s. com csv.firstLine(line); line = in.readLine(); int correct = 0; int wrong = 0; while (line != null) { Vector v = new SequentialAccessSparseVector(lmp.getNumFeatures()); int target = csv.processLine(line, v); System.out.println(line); String[] split = line.split(","); double score = lr.classifyFull(v).maxValueIndex(); if (score == target) correct++; else wrong++; System.out.println("Target is: " + target + " Score: " + score); line = in.readLine(); collector.add(target, score); } double posto = ((double) wrong / (double) (correct + wrong)) * 100; System.out.println("Total: " + (correct + wrong) + " Correct: " + correct + " Wrong: " + wrong + " Wrong pct: " + posto + "%"); //PrintWriter output = null; Matrix m = collector.confusion(); //output.printf(Locale.ENGLISH, "confusion: [[%.1f, %.1f], [%.1f, %.1f]]%n",m.get(0, 0), m.get(1, 0), m.get(0, 1), m.get(1, 1)); System.out.println("Confusion:" + m.get(0, 0) + " " + m.get(1, 0) + "\n \t " + m.get(0, 1) + " " + m.get(1, 1) + " "); // m = collector.entropy(); //output.printf(Locale.ENGLISH, "entropy: [[%.1f, %.1f], [%.1f, %.1f]]%n",m.get(0, 0), m.get(1, 0), m.get(0, 1), m.get(1, 1)); }
From source file:OpioidePrescriberClassification.Driver.java
public static void main(String args[]) throws Exception { List<Opioides> calls = Lists.newArrayList(new Parser("/input1/try.csv")); double heldOutPercentage = 0.10; // for (int run = 0; run < 20; run++) {//from www .j a v a2 s . co m // Random random = RandomUtils.getRandom(); Collections.shuffle(calls); int cutoff = (int) (heldOutPercentage * calls.size()); List<Opioides> test = calls.subList(0, cutoff); List<Opioides> train = calls.subList(cutoff, calls.size()); OnlineLogisticRegression lr = new OnlineLogisticRegression(NUM_CATEGORIES, Opioides.FEATURES, new L1()) .learningRate(1).alpha(1).lambda(0.000001).stepOffset(10000).decayExponent(0.2); // for (int pass = 0; pass < 2 ; pass++) { System.err.println("pass"); for (Opioides observation : train) { lr.train(observation.getTarget(), observation.asVector()); } // if (pass % 2 == 0) { Auc eval = new Auc(0.5); for (Opioides testCall : test) { eval.add(testCall.getTarget(), lr.classifyScalar(testCall.asVector())); } System.out.printf("%d, %.4f, %.4f\n", 1, lr.currentLearningRate(), eval.auc()); } } } }