{"id":17673850,"url":"https://github.com/yoyolicoris/pytorch-wise-ale","last_synced_at":"2025-05-12T19:56:10.862Z","repository":{"id":140051862,"uuid":"189543610","full_name":"yoyolicoris/pytorch-wise-ale","owner":"yoyolicoris","description":null,"archived":false,"fork":false,"pushed_at":"2019-06-21T13:01:43.000Z","size":589,"stargazers_count":5,"open_issues_count":1,"forks_count":1,"subscribers_count":3,"default_branch":"master","last_synced_at":"2025-05-11T21:52:09.485Z","etag":null,"topics":["kl-divergence","vae"],"latest_commit_sha":null,"homepage":null,"language":"Jupyter Notebook","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/yoyolicoris.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":"2019-05-31T06:54:07.000Z","updated_at":"2023-07-17T06:41:23.000Z","dependencies_parsed_at":null,"dependency_job_id":"71f8ef77-67a2-4100-8b6b-b63820e9146f","html_url":"https://github.com/yoyolicoris/pytorch-wise-ale","commit_stats":null,"previous_names":["yoyolicoris/pytorch-wise-ale"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/yoyolicoris%2Fpytorch-wise-ale","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/yoyolicoris%2Fpytorch-wise-ale/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/yoyolicoris%2Fpytorch-wise-ale/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/yoyolicoris%2Fpytorch-wise-ale/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/yoyolicoris","download_url":"https://codeload.github.com/yoyolicoris/pytorch-wise-ale/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":253813759,"owners_count":21968550,"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":["kl-divergence","vae"],"created_at":"2024-10-24T06:21:18.103Z","updated_at":"2025-05-12T19:56:10.830Z","avatar_url":"https://github.com/yoyolicoris.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Pytorch WiSE-ALE \nImplementation of [WiSE-ALE: Wide Sample Estimator for Approximate\nLatent Embedding](https://arxiv.org/abs/1902.06160).\nThe project structure is brought from [pytorch-template](https://github.com/victoresque/pytorch-template).\n\n## Quick start\n\nAn example config file `mnist.json`' is provided.\n\n~~~~\n{\n    \"name\": \"MNIST\",\n    \"n_gpu\": 1,\n\n    \"arch\": {\n        \"type\": \"MNIST_VAE\",\n        \"args\": {}\n    },\n    \"data_loader\": {\n        \"type\": \"MnistDataLoader\",\n        \"args\":{\n            \"data_dir\": \"mnist_data/\",\n            \"batch_size\": 64,\n            \"shuffle\": true,\n            \"validation_split\": 0,\n            \"num_workers\": 1\n        }\n    },\n    \"optimizer\": {\n        \"type\": \"Adam\",\n        \"args\":{\n            \"lr\": 0.001\n        }\n    },\n    \"loss\": \"WiSE_UB2\",     // Default loss function is WiSE-UB. Type 'AEVB' will use vanilla VAE objective.\n    \"metrics\": [\n        \"kl_div\", \"reconstruct\"\n    ],\n    \"lr_scheduler\": {\n        \"type\": \"StepLR\",\n        \"args\": {\n            \"step_size\": 60,\n            \"gamma\": 0.5\n        }\n    },\n    \"trainer\": {\n        \"epochs\": 30,\n        \"save_dir\": \"mnist_saved/\",\n        \"save_period\": 10,\n        \"verbosity\": 2,\n        \"monitor\": \"off\",\n        \"early_stop\": 0,\n\n        \"tensorboardX\": true,\n        \"sample_size\": 2\n    }\n}\n~~~~\nThe setting is the same as in the paper appendix.\n\nTo train the model with example config:\n```\npython train.py -c mnist.json\n```\nThe checkpoint files will be saved in `mnist_saved`. Other instructions please refer to pytorch-template.\n\n## Example Files\n\n* [mnist_visualization.ipynb](mnist_visualization.ipynb): A latent embedding visualization on mnist dataset.","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fyoyolicoris%2Fpytorch-wise-ale","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fyoyolicoris%2Fpytorch-wise-ale","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fyoyolicoris%2Fpytorch-wise-ale/lists"}