How to get the distribution of a doc over topics (and topic over words)

[RamtinYazdanian]

Hello,

First of all, thanks for developing this for Python!

I have been looking at the code and I cannot seem to find a way to infer the distribution of a document over the topics in its path from the root to the leaf (which would be the parameter theta in the "Hierarchical Topic Models and the Nested Chinese Restaurant Process" paper) and also the distribution of a topic over the words (which would be betas in the same paper).

For the second case, dividing word counts at a node by the sum of word counts should yield the probabilities of the respective topic over the words, but is that the best approximation of those values or is there a way to get a more accurate one?

[rana-alshaikh]

I have the same question, did you find the answer?

[gauravkoradiya]

same question here.