build based on caf87f6

dev/.documenter-siteinfo.json

@@ -1 +1 @@
-{"documenter":{"julia_version":"1.10.4","generation_timestamp":"2024-07-23T19:27:47","documenter_version":"1.5.0"}}
+{"documenter":{"julia_version":"1.10.5","generation_timestamp":"2024-10-07T18:14:25","documenter_version":"1.7.0"}}

dev/index.html

@@ -49,4 +49,4 @@
 decoder = AutoEncoderToolkit.SimpleGaussianDecoder(deconv_layers)
 
 # Re-defining the variational autoencoder
-vae = encoder * decoder</code></pre><p>Everything else in our training pipeline would remain the same thanks to multiple dispatch.</p><p>Furthermore, let&#39;s say that we would like to use a different flavor for our variational autoencoder. In particular the <code>InfoVAE</code> (also known as <code>MMD-VAE</code>) includes extra terms in the loss function to maximize mutual information between the latent space and the input data. We can easily take our <code>vae</code> model and convert it into a <code>MMDVAE</code>-type object from the <code>MMDVAEs</code> submodule as follows:</p><pre><code class="language-julia hljs">mmdvae = AutoEncoderToolkit.MMDVAEs.MMDVAE(vae)</code></pre><p>This is the power of <code>AutoEncoderToolkit.jl</code> and Julia&#39;s multiple dispatch!</p><h2 id="Implemented-Autoencoders"><a class="docs-heading-anchor" href="#Implemented-Autoencoders">Implemented Autoencoders</a><a id="Implemented-Autoencoders-1"></a><a class="docs-heading-anchor-permalink" href="#Implemented-Autoencoders" title="Permalink"></a></h2><table><tr><th style="text-align: right">model</th><th style="text-align: right">module</th><th style="text-align: right">description</th></tr><tr><td style="text-align: right">Autoencoder</td><td style="text-align: right"><a href="ae/#AEsmodule"><code>AEs</code></a></td><td style="text-align: right">Vanilla deterministic autoencoder</td></tr><tr><td style="text-align: right">Variational Autoencoder</td><td style="text-align: right"><a href="vae/#VAEsmodule"><code>VAEs</code></a></td><td style="text-align: right">Vanilla variational autoencoder</td></tr><tr><td style="text-align: right">β-VAE</td><td style="text-align: right"><a href="vae/#VAEsmodule"><code>VAEs</code></a></td><td style="text-align: right">beta-VAE to weigh the reconstruction vs. KL divergence in ELBO</td></tr><tr><td style="text-align: right">MMD-VAEs</td><td style="text-align: right"><a href="mmdvae/#MMDVAEsmodule"><code>MMDs</code></a></td><td style="text-align: right">Maximum-Mean Discrepancy Variational Autoencoders</td></tr><tr><td style="text-align: right">InfoMax-VAEs</td><td style="text-align: right"><a href="infomaxvae/#InfoMaxVAEsmodule"><code>InfoMaxVAEs</code></a></td><td style="text-align: right">Information Maximization Variational Autoencoders</td></tr><tr><td style="text-align: right">Hamiltonian VAE</td><td style="text-align: right"><a href="hvae/#HVAEsmodule"><code>HVAEs</code></a></td><td style="text-align: right">Hamiltonian Variational Autoencoders</td></tr><tr><td style="text-align: right">Riemannian Hamiltonian-VAE</td><td style="text-align: right"><a href="rhvae/#RHVAEsmodule"><code>RHVAEs</code></a></td><td style="text-align: right">Riemannian-Hamiltonian Variational Autoencoder</td></tr></table><div class="admonition is-success"><header class="admonition-header">Looking for contributors!</header><div class="admonition-body"><p>If you are interested in contributing to the package to add a new model, please check the <a href="https://github.com/mrazomej/AutoEncoderToolkit.jl">GitHub repository</a>. We are always  looking to expand the list of available models. And <code>AutoEncoderToolkit.jl</code>&#39;s  structure should make it relatively easy.</p></div></div><h2 id="GPU-support"><a class="docs-heading-anchor" href="#GPU-support">GPU support</a><a id="GPU-support-1"></a><a class="docs-heading-anchor-permalink" href="#GPU-support" title="Permalink"></a></h2><p><code>AutoEncoderToolkit.jl</code> supports GPU training out of the box for <code>CUDA.jl</code>-compatible GPUs. The <code>CUDA</code> functionality is provided as an extension. Therefore, to train a model on the GPU, simply import <code>CUDA</code> into the current environment, then move the model and data to the GPU. The rest of the training pipeline remains the same.</p></article><nav class="docs-footer"><a class="docs-footer-nextpage" href="quickstart/">Quick Start »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="auto">Automatic (OS)</option><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="catppuccin-latte">catppuccin-latte</option><option value="catppuccin-frappe">catppuccin-frappe</option><option value="catppuccin-macchiato">catppuccin-macchiato</option><option value="catppuccin-mocha">catppuccin-mocha</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.5.0 on <span class="colophon-date" title="Tuesday 23 July 2024 19:27">Tuesday 23 July 2024</span>. Using Julia version 1.10.4.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>
+vae = encoder * decoder</code></pre><p>Everything else in our training pipeline would remain the same thanks to multiple dispatch.</p><p>Furthermore, let&#39;s say that we would like to use a different flavor for our variational autoencoder. In particular the <code>InfoVAE</code> (also known as <code>MMD-VAE</code>) includes extra terms in the loss function to maximize mutual information between the latent space and the input data. We can easily take our <code>vae</code> model and convert it into a <code>MMDVAE</code>-type object from the <code>MMDVAEs</code> submodule as follows:</p><pre><code class="language-julia hljs">mmdvae = AutoEncoderToolkit.MMDVAEs.MMDVAE(vae)</code></pre><p>This is the power of <code>AutoEncoderToolkit.jl</code> and Julia&#39;s multiple dispatch!</p><h2 id="Implemented-Autoencoders"><a class="docs-heading-anchor" href="#Implemented-Autoencoders">Implemented Autoencoders</a><a id="Implemented-Autoencoders-1"></a><a class="docs-heading-anchor-permalink" href="#Implemented-Autoencoders" title="Permalink"></a></h2><table><tr><th style="text-align: right">model</th><th style="text-align: right">module</th><th style="text-align: right">description</th></tr><tr><td style="text-align: right">Autoencoder</td><td style="text-align: right"><a href="ae/#AEsmodule"><code>AEs</code></a></td><td style="text-align: right">Vanilla deterministic autoencoder</td></tr><tr><td style="text-align: right">Variational Autoencoder</td><td style="text-align: right"><a href="vae/#VAEsmodule"><code>VAEs</code></a></td><td style="text-align: right">Vanilla variational autoencoder</td></tr><tr><td style="text-align: right">β-VAE</td><td style="text-align: right"><a href="vae/#VAEsmodule"><code>VAEs</code></a></td><td style="text-align: right">beta-VAE to weigh the reconstruction vs. KL divergence in ELBO</td></tr><tr><td style="text-align: right">MMD-VAEs</td><td style="text-align: right"><a href="mmdvae/#MMDVAEsmodule"><code>MMDs</code></a></td><td style="text-align: right">Maximum-Mean Discrepancy Variational Autoencoders</td></tr><tr><td style="text-align: right">InfoMax-VAEs</td><td style="text-align: right"><a href="infomaxvae/#InfoMaxVAEsmodule"><code>InfoMaxVAEs</code></a></td><td style="text-align: right">Information Maximization Variational Autoencoders</td></tr><tr><td style="text-align: right">Hamiltonian VAE</td><td style="text-align: right"><a href="hvae/#HVAEsmodule"><code>HVAEs</code></a></td><td style="text-align: right">Hamiltonian Variational Autoencoders</td></tr><tr><td style="text-align: right">Riemannian Hamiltonian-VAE</td><td style="text-align: right"><a href="rhvae/#RHVAEsmodule"><code>RHVAEs</code></a></td><td style="text-align: right">Riemannian-Hamiltonian Variational Autoencoder</td></tr></table><div class="admonition is-success"><header class="admonition-header">Looking for contributors!</header><div class="admonition-body"><p>If you are interested in contributing to the package to add a new model, please check the <a href="https://github.com/mrazomej/AutoEncoderToolkit.jl">GitHub repository</a>. We are always  looking to expand the list of available models. And <code>AutoEncoderToolkit.jl</code>&#39;s  structure should make it relatively easy.</p></div></div><h2 id="GPU-support"><a class="docs-heading-anchor" href="#GPU-support">GPU support</a><a id="GPU-support-1"></a><a class="docs-heading-anchor-permalink" href="#GPU-support" title="Permalink"></a></h2><p><code>AutoEncoderToolkit.jl</code> supports GPU training out of the box for <code>CUDA.jl</code>-compatible GPUs. The <code>CUDA</code> functionality is provided as an extension. Therefore, to train a model on the GPU, simply import <code>CUDA</code> into the current environment, then move the model and data to the GPU. The rest of the training pipeline remains the same.</p></article><nav class="docs-footer"><a class="docs-footer-nextpage" href="quickstart/">Quick Start »</a><div class="flexbox-break"></div><p class="footer-message">Powered by <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> and the <a href="https://julialang.org/">Julia Programming Language</a>.</p></nav></div><div class="modal" id="documenter-settings"><div class="modal-background"></div><div class="modal-card"><header class="modal-card-head"><p class="modal-card-title">Settings</p><button class="delete"></button></header><section class="modal-card-body"><p><label class="label">Theme</label><div class="select"><select id="documenter-themepicker"><option value="auto">Automatic (OS)</option><option value="documenter-light">documenter-light</option><option value="documenter-dark">documenter-dark</option><option value="catppuccin-latte">catppuccin-latte</option><option value="catppuccin-frappe">catppuccin-frappe</option><option value="catppuccin-macchiato">catppuccin-macchiato</option><option value="catppuccin-mocha">catppuccin-mocha</option></select></div></p><hr/><p>This document was generated with <a href="https://github.com/JuliaDocs/Documenter.jl">Documenter.jl</a> version 1.7.0 on <span class="colophon-date" title="Monday 7 October 2024 18:14">Monday 7 October 2024</span>. Using Julia version 1.10.5.</p></section><footer class="modal-card-foot"></footer></div></div></div></body></html>