{"id":16833728,"url":"https://github.com/alecokas/flux-vae","last_synced_at":"2025-04-11T04:33:28.146Z","repository":{"id":52111581,"uuid":"275247380","full_name":"alecokas/flux-vae","owner":"alecokas","description":"Implementation of a Convolutional Variational Autoencoder in Flux.jl ","archived":false,"fork":false,"pushed_at":"2024-01-22T17:04:07.000Z","size":88,"stargazers_count":13,"open_issues_count":1,"forks_count":5,"subscribers_count":1,"default_branch":"master","last_synced_at":"2025-03-25T02:43:29.330Z","etag":null,"topics":["blog","cvae","fashion-mnist","flux","julia","julialang","machine-learning","vae","variational-autoencoder"],"latest_commit_sha":null,"homepage":"http://alecokas.github.io/julia/flux/vae/2020/07/22/convolutional-vae-in-flux.html","language":"Julia","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/alecokas.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":"2020-06-26T21:04:10.000Z","updated_at":"2024-05-22T12:23:39.000Z","dependencies_parsed_at":"2025-02-18T18:35:10.018Z","dependency_job_id":"46c35faf-08fa-4321-9518-13d098ff39b8","html_url":"https://github.com/alecokas/flux-vae","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/alecokas%2Fflux-vae","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/alecokas%2Fflux-vae/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/alecokas%2Fflux-vae/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/alecokas%2Fflux-vae/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/alecokas","download_url":"https://codeload.github.com/alecokas/flux-vae/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":248345203,"owners_count":21088231,"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":["blog","cvae","fashion-mnist","flux","julia","julialang","machine-learning","vae","variational-autoencoder"],"created_at":"2024-10-13T11:55:05.755Z","updated_at":"2025-04-11T04:33:28.079Z","avatar_url":"https://github.com/alecokas.png","language":"Julia","funding_links":[],"categories":[],"sub_categories":[],"readme":"# flux-vae\nThis repository accompanies my blog post [\"Convolutional VAE in Flux\"](http://alecokas.github.io/julia/flux/vae/2020/07/22/convolutional-vae-in-flux.html) where we take a look at variational autoencoders and do a walk-through demo of a [Flux](https://fluxml.ai/) implementation of a convolutional VAE using the FashionMNIST benchmark dataset.\n\n\u003cdiv style=\"text-align:center\"\u003e\u003cimg src=\"/images/vae.png\" width=\"600\"/\u003e\n\u003c/div\u003e\u003cbr/\u003e\n\n## Training\nYou can train the model directly from these scripts by running the following in your terminal:\n```\njulia --project=vaeenv conv-vae/main.jl\n```\nYou should see a progress tracker and the training loss displayed in your terminal.\n\n## Visualisation\nSimilarly, you can visualise reconstructed images from the test set by running:\n```\njulia --project=vaeenv conv-vae/visualise.jl\n```\nSome original test set samples (on the left) with the coresponding reconstructions (on the right):\n\n\u003cimg src=\"/images/recon.png\" alt=\"reconstruction-drawing\" width=\"400\"/\u003e\n\n## Dependencies\nThis code has been tested using Julia version 1.4.1. The package environment status is as follows:\n```\n  [fbb218c0] BSON v0.2.6\n  [336ed68f] CSV v0.7.1\n  [3895d2a7] CUDAapi v4.0.0\n  [35d6a980] ColorSchemes v3.9.0\n  [3a865a2d] CuArrays v2.2.0\n  [a93c6f00] DataFrames v0.21.4\n  [31c24e10] Distributions v0.23.2\n  [ced4e74d] DistributionsAD v0.5.2\n  [587475ba] Flux v0.10.4\n  [6a3955dd] ImageFiltering v0.6.13\n  [82e4d734] ImageIO v0.2.0\n  [916415d5] Images v0.22.2\n  [c8e1da08] IterTools v1.3.0\n  [eb30cadb] MLDatasets v0.5.2\n  [442fdcdd] Measures v0.3.1\n  [a3a9e032] NIfTI v0.4.1\n  [d96e819e] Parameters v0.12.1\n  [91a5bcdd] Plots v1.3.5\n  [92933f4c] ProgressMeter v1.3.1\n  [e88e6eb3] Zygote v0.4.20\n  [9a3f8284] Random \n  [10745b16] Statistics \n```\nYou can also have a look in the `vaeenv` directory where I have commited my `Project.toml` and `Manifest.toml` files for you to inspect the dependences.\n\n\nTo cite this work, please site the linked blog post:\n\u003cdiv class=\"language-plaintext highlighter-rouge\"\u003e\u003cdiv class=\"highlight\"\u003e\u003cpre class=\"highlight\"\u003e\n\u003ccode\u003e@article{kastanos20fluxvae,\n  title   = \"Convolutional VAE in Flux\",\n  author  = \"Alexandros Kastanos\",\n  journal = \"alecokas.github.io\",\n  year    = \"2020\",\n  url     = \"http://alecokas.github.io/julia/flux/vae/2020/07/22/convolutional-vae-in-flux.html\"\n}\n\u003c/code\u003e\u003c/pre\u003e\u003c/div\u003e\u003c/div\u003e\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Falecokas%2Fflux-vae","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Falecokas%2Fflux-vae","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Falecokas%2Fflux-vae/lists"}