How to convert a sparse matrix in the form of a list of lists of tuples to a dense matrix?

Question

numpy

There is an LDA model from the gensim.models.ldamodel module. Its get_document_topics(corpus) method returns an object that can be converted to a list of lists of tuples: [[(0, 0.09714252),
(1, 0.04680671), ..., where the length of the list is the document axis and the first number of the tuple is the topic axis , and the second is the probability of the subject of the document.

I would like to convert this list into a dense numpy array. Are there ready-made solutions?

So far I'm doing this (I don't like it when there are a lot of loops in the code):

document_topics = pd.concat(
    [pd.Series(dict(doc)) for doc in lda.get_document_topics(corpus)],
    axis=1
    ).T.fillna(0).values

Please tell me a more elegant solution, preferably in one line.