text mining - Perplexity in topic modeling -

i have run lda using topic models package on training data. how can determine perplexity of fitted model? read instruction, not sure code should use.

here's have far:

burnin <- 500 iter <- 1000 #keep <- 30 k <- 4  results_training <- lda(dtm_training, k,                 method = "gibbs",                 control = list(burnin = burnin,                                iter = iter))  terms <- terms(results_training, 10)  topic <- topics(results_training, 4)  # posterior probability each document on each topic posterior <- posterior(results_training)[[2]] 

it works perfectly, question how can use perplexity on testing data (results_testing)? , how can interpret result of perplexity?

thanks