python - Latent Dirichlet Allocation using Gensim on more than one corpus -
i have 2 questions related usage of gensim lda.
1) how can create model using 1 corpus, save , perhaps extend later on corpus training model on ? possible ?
2) can lda used classify unseen document, or model needs created again including in corpus ? there online way , see changes on fly ?
i have basic understanding of lda , have used topic modeling on simple corpus using lda , gensim libraries. please point out conceptual inconsistencies in question. !
i found helpful. gensim allow corpus added(updated) existing lda model. module allows both lda model estimation training corpus , inference of topic distribution on new, unseen documents. described here -
https://radimrehurek.com/gensim/models/ldamodel.html
additionally, algorithm streamed , can process corpora larger ram. has multicore implementation speed process.
lda = ldamodel(corpus, num_topics=10) lda.update(other_corpus)
this how model can updated.
Comments
Post a Comment