public void trainDocuments(
InstanceList documents, int numTopics, int numIterations, double alpha, double beta) {
double alphaSum = alpha * numTopics;
lda = new ParallelTopicModel(numTopics, alphaSum, beta);
lda.addInstances(documents);
lda.setNumThreads(4);
lda.setNumIterations(numIterations);
lda.printLogLikelihood = false;
try {
lda.estimate();
} catch (IOException e) {
// TODO Auto-generated catch block
e.printStackTrace();
}
System.out.println("LDA parameter, alphaSum: " + lda.alphaSum + ", beta: " + lda.beta);
}
public static void main(String[] args) throws Exception {
InstanceList instances = InstanceList.load(new File(args[0]));
int numTopics = Integer.parseInt(args[1]);
ParallelTopicModel model = new ParallelTopicModel(numTopics, 5.0, 0.01);
model.addInstances(instances);
model.setNumIterations(1000);
model.estimate();
TopicModelDiagnostics diagnostics = new TopicModelDiagnostics(model, 20);
if (args.length == 3) {
PrintWriter out = new PrintWriter(args[2]);
out.println(diagnostics.toXML());
out.close();
}
}
