https://github.com/oneflow-inc/zhusuan-oneflow
Zhusuan with backend Oneflow
https://github.com/oneflow-inc/zhusuan-oneflow
Last synced: about 2 months ago
JSON representation
Zhusuan with backend Oneflow
- Host: GitHub
- URL: https://github.com/oneflow-inc/zhusuan-oneflow
- Owner: Oneflow-Inc
- License: mit
- Created: 2021-07-03T13:38:20.000Z (almost 5 years ago)
- Default Branch: main
- Last Pushed: 2021-07-05T04:03:26.000Z (almost 5 years ago)
- Last Synced: 2025-01-01T18:27:39.790Z (over 1 year ago)
- Language: Python
- Size: 52.7 KB
- Stars: 1
- Watchers: 7
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Zhusuan-Oneflow
This work is based on the repo: [Zhusuan-PaddlePaddle](https://github.com/McGrady00H/Zhusuan-PaddlePaddle).
ZhuSuan-Oneflow is a python probabilistic programming library for
**Bayesian deep learning**, which conjoins the complimentary advantages of
Bayesian methods and deep learning. ZhuSuan is built upon
[Oneflow](https://github.com/Oneflow-Inc/oneflow). Unlike existing deep learning
libraries, which are mainly designed for deterministic neural networks and
supervised tasks, ZhuSuan-Oneflow provides deep learning style primitives and
algorithms for building probabilistic models and applying Bayesian inference.
The supported inference algorithms include:
* Variational inference with programmable variational posteriors, various
objectives and advanced gradient estimators (SGVB, SWI, etc.).
* Importance sampling for learning and evaluating models, with programmable
proposals.
* MCMC samplers: Hamiltonian Monte Carlo (HMC) with parallel chains, and
Stochastic Gradient MCMC (sgmcmc).
## Installation
ZhuSuan-Oneflow is still under development.
### Install Oneflow(version >= 0.4.0)
https://github.com/Oneflow-Inc/oneflow#install-with-pip-package
### Install requirements