public static void main(String[] args) throws Exception {
int minTokenCount = 5;
File corpusFile = new File(args[0]);
CharSequence[] articleTexts = LdaWormbase.readCorpus(corpusFile);
SymbolTable symbolTable = new MapSymbolTable();
TokenizerFactory tokenizerFactory = LdaWormbase.WORMBASE_TOKENIZER_FACTORY;
int[][] docTokens =
LatentDirichletAllocation.tokenizeDocuments(
articleTexts, tokenizerFactory, symbolTable, minTokenCount);
LdaRunnable runnable1 =
new LdaRunnable(docTokens, new LdaReportingHandler(symbolTable), new Random());
LdaRunnable runnable2 =
new LdaRunnable(docTokens, new LdaReportingHandler(symbolTable), new Random());
Thread thread1 = new Thread(runnable1);
Thread thread2 = new Thread(runnable2);
thread1.start();
thread2.start();
thread1.join();
thread2.join();
LatentDirichletAllocation lda0 = runnable1.mLda;
LatentDirichletAllocation lda1 = runnable2.mLda;
System.out.println("\nComputing Greedy Aligned Symmetrized KL Divergences");
double[] scores = similarity(lda0, lda1);
for (int i = 0; i < scores.length; ++i) System.out.printf("%4d %15.3f\n", i, scores[i]);
}
