This repository contains a PyTorch implementation of a diffusion model trained on a time series dataset. The diffusion model is a generative model that learns a latent representation of the data through a diffusion process. The code includes the definition of the diffusion model, the training procedure, and visualization tools for the generated samples.