Copy of [0.5] GANs & VAEs (exercises).ipynb - Colaboratory
Please send any problems / bugs on the #errata channel in the Slack group, and ask any questions on the dedicated channels for this chapter of material. Today, we're studying two important classes of generative image models: Generative Adversarial Networks (GANs) and Variational Autoencoders (VAEs). Although these generally aren't SOTA any more (thanks in part to the rise of diffusion models), there are some deep conceptual insights which can be gleaned from studying these models (VAEs in particular) which help lay the groundwork for more advanced models. These exercises will also hopefully bring much of this chapter full-circle: We start with some reading material on GANs and transposed convolutions (which are often used in parallel with convolutions, to take a latent space and map it back into a full-size image). Next, you'll actually implement and train your own GANs, to generate celebrity pictures. By the time you're done, you'll hopefully have produced output like this (and you'll
Google Colab Sign in
Explore this link on the map →related reading
- What are Diffusion Models? | Lil'Loglilianweng.github.io
- Replicate - Run AI with an APIreplicate.com
- Difference between AutoEncoder (AE) and Variational AutoEncoder (VAE) | Towards Data Sciencetowardsdatascience.com
- A Gentle Introduction to Generative Adversarial Networks (GANs) - MachineLearningMastery.commachinelearningmastery.com
- Generative modelling in latent space – Sander Dielemansander.ai
- Adversarial Autoencodersarxiv.org
- ⭐️ Diffusion Modelsandrewkchan.dev
- Yang Songyang-song.net
- francesco215.github.io/autoregressive_diffusion/francesco215.github.io
- Mode collapse - Wikipediaen.wikipedia.org
- State-of-the-Art Image Generative Models – Aran Komatsuzakiarankomatsuzaki.wordpress.com
- Jupyter Notebook Viewernbviewer.org