{"id":13400413,"url":"https://github.com/thu-ml/zhusuan","last_synced_at":"2025-05-15T09:06:31.483Z","repository":{"id":37678029,"uuid":"63605558","full_name":"thu-ml/zhusuan","owner":"thu-ml","description":"A probabilistic programming library for Bayesian deep learning, generative models, based on Tensorflow","archived":false,"fork":false,"pushed_at":"2022-12-17T20:33:19.000Z","size":1836,"stargazers_count":2219,"open_issues_count":13,"forks_count":419,"subscribers_count":143,"default_branch":"master","last_synced_at":"2025-04-14T15:00:40.892Z","etag":null,"topics":["bayesian-inference","deep-learning","generative-models","graphical-models","probabilistic-programming"],"latest_commit_sha":null,"homepage":"http://zhusuan.readthedocs.io","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/thu-ml.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":"CONTRIBUTING.md","funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null}},"created_at":"2016-07-18T13:31:38.000Z","updated_at":"2025-03-28T06:10:27.000Z","dependencies_parsed_at":"2023-01-29T17:55:20.129Z","dependency_job_id":null,"html_url":"https://github.com/thu-ml/zhusuan","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/thu-ml%2Fzhusuan","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/thu-ml%2Fzhusuan/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/thu-ml%2Fzhusuan/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/thu-ml%2Fzhusuan/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/thu-ml","download_url":"https://codeload.github.com/thu-ml/zhusuan/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":254310515,"owners_count":22049469,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["bayesian-inference","deep-learning","generative-models","graphical-models","probabilistic-programming"],"created_at":"2024-07-30T19:00:51.752Z","updated_at":"2025-05-15T09:06:31.467Z","avatar_url":"https://github.com/thu-ml.png","language":"Python","funding_links":[],"categories":["Python","Probabilistic Programming Languages (PPL) (a.k.a. Build your own Topic Model)","概率统计","\u003cspan id=\"head50\"\u003e3.6. Probablistic Machine Learning and Deep Learning\u003c/span\u003e"],"sub_categories":["Embedding based Topic Models","Semi-supervised Learning"],"readme":"\u003cdiv align=\"center\"\u003e\n  \u003ca href=\"http://zhusuan.readthedocs.io\"\u003e \u003cimg width=\"250px\" height=\"auto\" \n  src=\"docs/_static/images/zhusuan-logo-v.png\"\u003e\u003c/a\u003e\n\u003c/div\u003e\n\n---\n\n[![Build Status](https://travis-ci.org/thu-ml/zhusuan.svg?branch=master)](https://travis-ci.org/thu-ml/zhusuan)\n[![Doc Status](https://readthedocs.org/projects/zhusuan/badge/?version=latest)](http://zhusuan.readthedocs.io/en/latest/?badge=latest)\n[![License](https://img.shields.io/badge/license-MIT-blue.svg)](https://github.com/thu-ml/zhusuan/blob/master/LICENSE)\n[![Join the chat at https://gitter.im/thu-ml/zhusuan](https://badges.gitter.im/thu-ml/zhusuan.svg)](https://gitter.im/thu-ml/zhusuan?utm_source=badge\u0026utm_medium=badge\u0026utm_campaign=pr-badge\u0026utm_content=badge)\n\n**ZhuSuan** is a Python probabilistic programming library for Bayesian deep\nlearning, which conjoins the complimentary advantages of Bayesian methods and\ndeep learning. ZhuSuan is built upon\n[TensorFlow](https://www.tensorflow.org). Unlike existing deep\nlearning libraries, which are mainly designed for deterministic neural\nnetworks and supervised tasks, ZhuSuan provides deep learning style primitives\nand algorithms for building probabilistic models and applying Bayesian\ninference. The supported inference algorithms include:\n\n* **Variational Inference (VI)** with programmable variational posteriors, various\n  objectives and advanced gradient estimators (SGVB, REINFORCE, VIMCO, etc.).\n\n* **Importance Sampling (IS)** for learning and evaluating models, with programmable\n  proposals.\n\n* **Hamiltonian Monte Carlo (HMC)** with parallel chains, and optional\n  automatic parameter tuning.\n\n* **Stochastic Gradient Markov Chain Monte Carlo (SGMCMC)**:\n  SGLD, PSGLD, SGHMC, and SGNHT.\n\n## Installation\n\nZhuSuan is still under development. Before the first stable release (1.0),\nplease clone the repository and run\n```\npip install .\n```\nin the main directory. This will install ZhuSuan and its dependencies\nautomatically. ZhuSuan also requires **TensorFlow 1.13.0 or later**. Because\nusers should choose whether to install the cpu or gpu version of TensorFlow,\nwe do not include it in the dependencies. See\n[Installing TensorFlow](https://www.tensorflow.org/install/).\n\nIf you are developing ZhuSuan, you may want to install in an\n\"editable\" or \"develop\" mode. Please refer to the Contributing section below.\n\n## Documentation\n\n* [Tutorials and API docs](http://zhusuan.readthedocs.io)\n* [ZhuSuan's white paper](https://arxiv.org/abs/1709.05870)\n\n## Examples\n\nWe provide examples on traditional hierarchical Bayesian models and recent\ndeep generative models.\n\nTo run the provided examples, you may need extra dependencies to be installed.\nThis can be done by\n```\npip install \".[examples]\"\n```\n* Gaussian: \n  [HMC](examples/toy_examples/gaussian.py)\n* Toy 2D Intractable Posterior:\n  [SGVB](examples/toy_examples/toy2d_intractable.py)\n* Bayesian Neural Networks:\n  [SGVB](examples/bayesian_neural_nets/bnn_vi.py),\n  [SGMCMC](examples/bayesian_neural_nets/bnn_sgmcmc.py)\n* Variational Autoencoder (VAE):\n  [SGVB](examples/variational_autoencoders/vae.py),\n  [IWAE](examples/variational_autoencoders/iwae.py)\n* Convolutional VAE:\n  [SGVB](examples/variational_autoencoders/vae_conv.py)\n* Semi-supervised VAE (Kingma, 2014):\n  [SGVB](examples/semi_supervised_vae/vae_ssl.py),\n  [Adaptive IS](examples/semi_supervised_vae/vae_ssl_adaptive_is.py)\n* Deep Sigmoid Belief Networks\n  [Adaptive IS](examples/sigmoid_belief_nets/sbn_adaptive_is.py),\n  [VIMCO](examples/sigmoid_belief_nets/sbn_vimco.py)\n* Logistic Normal Topic Model:\n  [HMC](examples/topic_models/lntm_mcem.py)\n* Probabilistic Matrix Factorization:\n  [HMC](examples/probabilistic_matrix_factorization/pmf_hmc.py)\n* Sparse Variational Gaussian Process:\n  [SGVB](examples/gaussian_process/svgp.py)\n\n## Citing ZhuSuan\n\nIf you find ZhuSuan useful, please cite it in your publications.\nWe provide a BibTeX entry of the ZhuSuan white paper below.\n```\n@ARTICLE{zhusuan2017,\n    title={Zhu{S}uan: A Library for {B}ayesian Deep Learning},\n    author={Shi, Jiaxin and Chen, Jianfei. and Zhu, Jun and Sun, Shengyang\n    and Luo, Yucen and Gu, Yihong and Zhou, Yuhao},\n    journal={arXiv preprint arXiv:1709.05870},\n    year=2017,\n}\n```\n\n## Contributing\n\nWe always welcome contributions to help make ZhuSuan better. If you would like \nto contribute, please check out the guidelines [here](CONTRIBUTING.md).\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fthu-ml%2Fzhusuan","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fthu-ml%2Fzhusuan","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fthu-ml%2Fzhusuan/lists"}