Latent Dirichlet Allocation Perplexity

Question

I am using MlLib implementation of Latent Dirichlet Allocation but I am having some trouble in understanding how perplexity is calculated. Based on the example provided by Apache logPerplexity is calculated on the same dataset on which the model is trained while in the documentation is noted that the input for model.logPerplexity must be a test set. Can anyone help me? Here is the code to which I refer:

// Loads data.
Dataset<Row> dataset = spark.read().format("libsvm")
  .load("data/mllib/sample_lda_libsvm_data.txt");

// Trains a LDA model.
LDA lda = new LDA().setK(10).setMaxIter(10);
LDAModel model = lda.fit(dataset);

double ll = model.logLikelihood(dataset);
double lp = model.logPerplexity(dataset);
System.out.println("The lower bound on the log likelihood of the entire corpus: " + ll);
System.out.println("The upper bound on perplexity: " + lp);

// Describe topics.
Dataset<Row> topics = model.describeTopics(3);
System.out.println("The topics described by their top-weighted terms:");
topics.show(false);

// Shows the result.
Dataset<Row> transformed = model.transform(dataset);
transformed.show(false);