This repository contains the implementation for the paper Information Theoretic StructuredGenerative Modeling,
Specially thanks for the open-source codes shared by sagelywizard/pytorch-mdn and PyTorch-GAN
- Pytorch
- A GPU Machine
- GAN Example
- MINE Example
- Density Estimation
- Conditonal Estimation
To run ood_visualization.ipynb, please download the pretrained model in the ./model/ folder. (google drive link later)
python MDN.py
python CGAN.pyGMM_VBGMM_CE.py provides codes for producing conditional CE for any mixture models obtained from scipy.
Simply calling
compute_conditionalCE(joint, gm_joint)in python to obtain the value, where joint should be bs*K and gm_joint is the class obtained from scipy.