Minchul Lee

Hi @PearlOnyx08 Thank you for your good suggestion. You can use `mdl.infer()` to calculate holdout perplexity, but it is somewhat cumbersome. In my opinion, even adding a method for holdout...

Hi @lucasgautheron , A detailed discussion related to your question is here(#147). Currently there are three options for calculating perplexity in tomotopy (`mdl.perplexity`, `mdl.infer()`, `doc.get_ll()`), but they are using subtly...

@DongqingSun96, Oops, sorry for late answer. I forgot this issue totally. Currently, `tomotopy` doesn't provide such function, because its internal implementation cannot accept a matrix in bag-of-words format. To insert...

Adding new features into `tomotopy.utils.Corpus` constructing from matrix: * `Corpus.from_dense_matrix(matrix, vocab_dict) -> Corpus` * `Corpus.from_sparse_matrix(matrix, vocab_dict) -> Corpus` new features constructing matrix from Corpus: * `Corpus.to_dense_matrix(self) -> numpy.ndarray` * `Corpus.to_sparse_matrix(self)...

Hello @Sixy1204 What does the word embedding vectors in topic space mean? After training, LDA model generates document-topic distributions and topic-word distributions. If you want to get topic distributions for...

@Sixy1204 You can normalize the value to make its sum to be 1 as follows: ```python topic_word_dists = np.array([mdl.get_topic_word_dist(k, normalize=False) for k in range(mdl.k)]) # turn off normalizing over word...

Duplicated issue: #153 . Current tomotopy doesn't support M1 and porting for M1 is working on. It can take a long time to work, so I recommend using Rosetta.

Hi @ebergam, sorry for the late answer. Thanks for the good suggestion. I'll take a look at the paper and see if it can be implemented efficiently.

Hi @ebergam Thank you for great suggestion. I'll review the paper and see if it can be added to the next major update.

Hi @batmanscode, Unfortunately there is no way to weight docs currently. Actually, I have conducted several experiments with different doc weightings before, but they didn't show any improvement compared to...