Structural Topic Modeling (or alternatives) for python

Question

I would like to run a Structural Topic Model (STM) on my data in python. I saw in this thread (now closed I think) that STM were still not implemented in gensim or other python libraries https://github.com/RaRe-Technologies/gensim/issues/1038

Does anyone know if STM's have been implemented since?

The alternative suggested are correlated topic models (CTM) available in tomatopy . Can one do the same thing with CTM's than with STM's or only in special cases? For example, can CTM's consider both topic correlation and metadata such as time? I am wondering if I should go through the trouble of learning R to use STM's or if there is a good alternative in python?

Answer 1

You could use Latent Dirichlet Allocation, which is a very similar approach. Here is a tutorial .