Home

Awesome

MoPoE-VAE

This is the official code for the ICLR 2021 paper "Generalized Multimodal ELBO". Here is the link to the OpenReview-Site: https://openreview.net/forum?id=5Y21V0RDBV

If you have any questions about the code or the paper, we are happy to help!

Preliminaries

This code was developed and tested with:

First, set up the conda enviroment as follows:

conda env create -f environment.yml  # create conda env
conda activate mopoe                 # activate conda env

Second, download the data, inception network, and pretrained classifiers:

curl -L -o tmp.zip https://drive.google.com/drive/folders/1lr-laYwjDq3AzalaIe9jN4shpt1wBsYM?usp=sharing
unzip tmp.zip
unzip celeba_data.zip -d data/
unzip data_mnistsvhntext.zip -d data/
unzip PolyMNIST.zip -d data/

Experiments

Experiments can be started by running the respective job_* script. To choose between running the MVAE, MMVAE, and MoPoE-VAE, one needs to change the script's METHOD variabe to "poe", "moe", or "joint_elbo" respectively. By default, each experiment uses METHOD="joint_elbo".

running MNIST-SVHN-Text

./job_mnistsvhntext

running PolyMNIST

./job_polymnist

running Bimodal Celeba

./job_celeba