davidjaw/diffusion-models-pytorch

Pytorch implementation of Diffusion Models (https://arxiv.org/pdf/2006.11239.pdf)

Diffusion Models

This is an easy-to-understand implementation of diffusion models within 100 lines of code. Different from other implementations, this code doesn't use the lower-bound formulation for sampling and strictly follows Algorithm 1 from the DDPM paper, which makes it extremely short and easy to follow.

---

Train a Diffusion Model on your own data:

Unconditional Training

1. (optional) Configure Hyperparameters in ddpm.py
2. Set path to dataset in ddpm.py
3. python ddpm.py

Sampling

The following examples show how to sample images using the models trained in the video on the Landscape Dataset. You can download the checkpoints for the models here. Examples also showed in inference.py file.

Unconditional Model

    device = "cuda"
    model = UNet().to(device)
    ckpt = torch.load("unconditional_ckpt.pt")
    model.load_state_dict(ckpt)
    diffusion = Diffusion(img_size=64, device=device)
    x = diffusion.sample(model, n=16)
    plot_images(x)

---

• This is a simplified version of the forked repo here
• A more advanced version of this code can be found here by @tcapelle. It introduces better logging, faster & more efficient training and other nice features and is also being followed by a nice write-up.