I am trying to fit a topic model using LDA with Gibbs sampling in python but I can’t find a package fitting my needs:
(1) Both LDAs in <scikit-learn> and <gensim}> don’t use Gibbs sampling;
(2) <Python wrapper for LDA from Mallet in JAVA in gensim> was removed and the latest release including it is version 3.8.3.

(1) I tried to install this downgraded version of gensim to use LdaMallet wrapper but a lot of other errors come from no where when i ran the programs;
(2) I also tried <lda> package but it doesn’t provide vocabulary control like gensim.corpora.dictionary.filter_extremes(). Neither does <little-mallet-wrapper> .

My basic need is to control vocabulary and fit LDA with Gibbs sampling in python. I have been looking for online for days but most of results are the implementations of LDA from scratch. Any advice is highly appreciated! Thanks