Autoencoding Variational Inference for Topic Models

Code for the ICLR 2017 paper: Autoencoding Variational Inference for Topic Models

> Arxiv

> OpenReview

Quick Start:

This is a tensorflow implementation for both of the Autoencoded Topic Models mentioned in the paper.

To run the prodLDA model in the 20Newgroup dataset:

CUDA_VISIBLE_DEVICES=0 python run.py -m prodlda -f 100 -s 100 -t 50 -b 200 -r 0.002 -e 80

Similarly for NVLDA:

CUDA_VISIBLE_DEVICES=0 python run.py -m nvlda -f 100 -s 100 -t 50 -b 200 -r 0.005 -e 300

Check run.py for other options.

UPDATE

Added topic_prop method to both the models. Softmax the output of this method to get the topic proportions.