https://github.com/mingxuan-yi/Bridging-the-gap-between-VI-and-WGF

https://github.com/mingxuan-yi/Bridging-the-gap-between-VI-and-WGF

Last synced: 11 days ago
JSON representation

• Host: GitHub
• URL: https://github.com/mingxuan-yi/Bridging-the-gap-between-VI-and-WGF
• Owner: mingxuan-yi
• Created: 2023-05-24T17:09:24.000Z (over 1 year ago)
• Default Branch: main
• Last Pushed: 2023-11-22T19:35:16.000Z (12 months ago)
• Last Synced: 2024-08-01T16:45:46.007Z (3 months ago)
• Language: Jupyter Notebook
• Size: 1.79 MB
• Stars: 3
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        

**First, setup the dependencies by**

>pip install -r requirements.txt

## 1.The Illustrative Example

see `notebook/The Illustrative Example.ipynb`.



## 2. The experiment used GMMs as the variational distribution.

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



## 3. The experiment on Bayesian logistic regressions.

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

## 4. To reproduce the experiment in Appendix B.1, run

>python 1d_gmm.py