Calculating topic distribution of an unseen document on GenSim

Question

I am trying to use LDA module of GenSim to do the following task

"Train a LDA model with one big document and keep track of 10 latent topics. Given a new, unseen document, predict probability distribution of 10 latent topics".

As per tutorial here: http://radimrehurek.com/gensim/tut2.html, this seems possible for a document in a corpus, but I am wondering if it it would be possible for an unseen document.

Thank you!

Answer 1

From the documentation you posted it looks like you can train your model like this:

>>> model = models.LdaModel(corpus, id2word=dictionary, num_topics=100)

And then from this page it looks like you can apply your model on "an unseen document" like this:

>>> doc_lda = model[doc_bow]

Where doc_bow is a bag-of-words generated by the doc2bow tool.