/pi-vae

Code for utilising VAE as means of doing exact MCMC inference in complex high-dimensional space

Primary LanguageJupyter NotebookMIT LicenseMIT

πVAE

Code for utilising VAE as means of doing exact MCMC inference in complex high-dimensional space.

Accompanying paper is πVAE: a stochastic process prior for Bayesian deep learning with MCMC

The πVAE model has 2 parts :

  1. Learning / encoding a stochastic prior via a VAE.
  2. Then using the learnt basis, and decoder network , perform inference on our data to get a posterior.

To run the code :

  1. Run src_py/models/pi_vae.py . To choose the the type of prior learnt, modify the training dataset, the current default is 1D GP.
  2. To perform inference using stan, use the file notebooks/pivae.stan by passing the model parameters learnt in the above step. An example is given in the notebooks, notebooks/pivae.stan and notebooks/run_monotonic_mcmc.ipynb