> ./paths _ > ./modules > ./articles > ./papers

# Diffusion Models

Dive into the math and code behind diffusion models, from score matching to building your own image generator.

└─ status: in-progress

2 sections · 3 modules

│ The mathematical foundations of diffusion

├─ 1.1 The Math of Diffusion

Forward and reverse diffusion processes, noise schedules, and the variational lower bound.

└─ 1.2 Score Matching and SDEs

Connect score-based generative models with stochastic differential equations for continuous diffusion.

math · theory · computer-vision

│ Building a diffusion model from scratch

└─ 2.1 Building a Tiny Diffusion Model

Implement a minimal DDPM from scratch in PyTorch and train it on a simple image dataset.

pytorch · hands-on · training · computer-vision