{"id":19509381,"url":"https://github.com/ondrejbiza/vae","last_synced_at":"2026-05-17T05:36:52.076Z","repository":{"id":111605855,"uuid":"167976102","full_name":"ondrejbiza/vae","owner":"ondrejbiza","description":"Variational autoencoders implemented in Tensorflow.","archived":false,"fork":false,"pushed_at":"2019-10-18T18:45:04.000Z","size":905,"stargazers_count":1,"open_issues_count":0,"forks_count":1,"subscribers_count":1,"default_branch":"master","last_synced_at":"2025-11-19T05:04:20.592Z","etag":null,"topics":["deep-learning","deep-neural-networks","tensorflow","variational-autoencoders","vq-vae"],"latest_commit_sha":null,"homepage":null,"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/ondrejbiza.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2019-01-28T14:30:40.000Z","updated_at":"2019-11-23T20:51:58.000Z","dependencies_parsed_at":null,"dependency_job_id":"21380e2d-75aa-4f39-811e-dc8ce6fd342f","html_url":"https://github.com/ondrejbiza/vae","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/ondrejbiza/vae","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ondrejbiza%2Fvae","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ondrejbiza%2Fvae/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ondrejbiza%2Fvae/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ondrejbiza%2Fvae/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/ondrejbiza","download_url":"https://codeload.github.com/ondrejbiza/vae/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ondrejbiza%2Fvae/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":33128676,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-05-16T18:38:32.183Z","status":"online","status_checked_at":"2026-05-17T02:00:05.366Z","response_time":107,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"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":["deep-learning","deep-neural-networks","tensorflow","variational-autoencoders","vq-vae"],"created_at":"2024-11-10T23:11:49.645Z","updated_at":"2026-05-17T05:36:52.060Z","avatar_url":"https://github.com/ondrejbiza.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Variational Autoencoders in Tensorflow\n\n\u003cp float=\"left\"\u003e\n    \u003cimg src=\"vae/results/vae_conv_samples.png\" alt=\"vae_mnist_samples\" width=\"420\"/\u003e\n    \u003cimg src=\"vae/results/vae_conv_losses.png\" alt=\"vae_mnist_losses\" width=\"420\"/\u003e\n\u003c/p\u003e\n\n## Set up\n\n* Install Python \u003e= 3.6.\n* Install packages in *requirements.txt*.\n* Tested with tensorflow-gpu 1.7.0 (CUDA 9.1, cuDNN 7.1) and tensorflow-gpu 1.14.0 (CUDA 10.0, cuDNN 7.6).\n* For tensorflow-gpu 1.14.0, use the flag --fix-cudnn if you get a cuDNN initialization error.\n## Usage\n\n### Autoencoder:\n```\n# ConvNet on MNIST\npython -m vae.scripts.ae_conv_mnist\n```\n\nMNIST, default settings: -54.26 test log-likelihood (1 run)\n\n### Variational Autoencoder (VAE):\n\n```\n# ConvNet on MNIST\npython -m vae.scripts.vae_conv_mnist\n\n# fully-connected net on MNIST\npython -m vae.scripts.vae_fc_mnist\n```\n\nPaper: https://arxiv.org/abs/1312.6114\n\nMNIST, ConvNet, default settings: -71.52 test log-likelihood (1 run)\n\n### VampPrior VAE:\n\n```\n# ConvNet on MNIST\npython -m vae.scripts.vampprior_vae_conv_mnist\n\n# fully-connected net on a toy dataset\npython -m vae.scripts.vampprior_vae_fc_toy\n```\n\nPaper: https://arxiv.org/abs/1705.07120\n\nMNIST, default settings: -70.08 test log-likelihood (1 run)\n\n### Gaussian Mixture Prior VAE:\n\n```\n# ConvNet on MNIST\npython -m vae.scripts.gmprior_vae_conv_mnist\n\n# fully-connected net on a toy dataset\npython -m vae.scripts.gmprior_vae_fc_toy\n```\n\nBaseline from https://arxiv.org/abs/1705.07120\n\nMNIST, ConvNet, default settings: -69.58 test log-likelihood (1 run)\n\n### Softmax-Gumbel VAE:\n\n```\n# ConvNet on MNIST\npython -m vae.scripts.sg_vae_conv_mnist\n```\n\nPaper: https://arxiv.org/abs/1611.01144\n\nMNIST, default settings: -81.56 test log-likelihood (1 run)\n\n### Vector Quantization VAE (VQ-VAE):\n\nMore or less a 1-on-1 copy of https://github.com/hiwonjoon/tf-vqvae/blob/master/model.py:\n```\npython -m vae.scripts.vq_vae_fully_conv_mnist\n```\n\nMy own version that seems to produce better samples:\n```\npython -m vae.scripts.vq_vae_conv_mnist\n```\n\nPaper: https://arxiv.org/abs/1711.00937\n\nI'm not sure how to measure the test log-likelihood here.\n\n## Notes\n\n* The architecture of all ConvNets is based on this paper (https://arxiv.org/abs/1803.10122) with half the filters.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fondrejbiza%2Fvae","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fondrejbiza%2Fvae","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fondrejbiza%2Fvae/lists"}