This repository provides introductory Jupyter notebooks for various deep generative models, targeting beginners and researchers. It aims to make complex concepts accessible and runnable on standard hardware, facilitating learning and experimentation in the field.

How It Works

The project implements a wide array of deep generative models, including Mixture of Gaussians, Autoregressive Models, Flow-based models, VAEs, Diffusion Models, Score-based Generative Models, Energy-based Models, GANs, and LLMs. Each example is designed to be simple and self-contained, allowing users to follow and run the code quickly.

Quick Start & Requirements

• Examples are provided as Jupyter notebooks.
• Requirements: pytorch 1.7.0, numpy 1.17.2, matplotlib 3.1.1, scikit-learn 0.21.3, pytorch-model-summary 0.1.1, jupyter 1.0.0.
• The project is designed to run on most laptops or computers.

Highlighted Details

• Covers 11 distinct categories of deep generative models.
• Examples are intentionally simplified for ease of understanding and rapid execution.
• Includes implementations for neural compression and a decoder-based transformer (teenyGPT).

Maintenance & Community

• The repository is associated with the book "Deep Generative Modeling" by Jakub M. Tomczak.
• Citation is requested via APA or BibTeX format, referencing the book.

Licensing & Compatibility

• The README does not explicitly state a license.

Limitations & Caveats

• The provided code versions (e.g., PyTorch 1.7.0) may be outdated, potentially requiring dependency updates for compatibility with newer environments.