public void saveModel() {
try {
File tmpFile = File.createTempFile("model", null);
ModelSerializer.writeModel(net.getNN(), tmpFile, true);
} catch (IOException ioe) {
}
}
public void clearAll() {
clearData();
try {
net.clear();
} catch (NullPointerException npe) {
Logger.getLogger(MLSegmentationService.class.getName()).log(Level.SEVERE, null, npe);
}
}
public INDArray classify() {
DataSetIterator iter = new ProfileIterator(testData, imgDatasets);
INDArray predict = net.output(iter);
// int size0 = predict.size(0);
// int size1 = predict.size(1);
// int size2 = predict.size(2);
// INDArray element = predict.getRow(0);
// System.out.println("predict DONE");
return predict;
}
public INDArray dummyClassification() {
DataSetIterator iter = new ProfileIterator(testData, imgDatasets);
INDArray predict = net.output(iter);
INDArray dummy =
Nd4j.create(new int[] {predict.size(0), predict.size(1), predict.size(2)}, 'f');
for (int i = 0; i < predict.size(0); i++) {
for (int j = 0; j < predict.size(2); j++) {
if (j < 10 || j > 14) dummy.putScalar(new int[] {i, 0, j}, 0.0);
else dummy.putScalar(new int[] {i, 0, j}, 1.0);
}
}
return dummy;
}
public void train() {
DataSetIterator iter = new ProfileIterator(trainingData, confirmationSet, imgDatasets, true);
int iEpoch = 0;
int nEpochs = 300;
while (iEpoch < nEpochs) {
System.out.printf("EPOCH %d\n", iEpoch);
Evaluation eval = new Evaluation();
while (iter.hasNext()) {
DataSet ds = iter.next();
net.train(ds);
INDArray predict2 = net.output(ds.getFeatureMatrix());
INDArray labels2 = ds.getLabels();
// eval.evalTimeSeries(labels2, predict2);
}
iter.reset();
// System.out.println(eval.stats());
iEpoch++;
}
System.out.println("Fitting : DONE");
}