mingxuan-yi / bridging-the-gap-between-vi-and-wgf Goto Github PK

First, setup the dependencies by

pip install -r requirements.txt

see notebook/The Illustrative Example.ipynb.

run python banana.py. Default parameters: number of mixture components --num_components=5, particles per component used for Monte Carlo gradients --num_sample=30.

run python train_Bayesian_logistic.py --dataset='ionos' --num_epoch=20001. Change --dataset to 'heart', 'pima', 'wine' for different UCI dataset.

python 1d_gmm.py