In our project, we trained a deep diffusion model (DDPM) by creating a simplified version of a U-Net architecture and using various datasets. This training process allowed our DDPM to learn how to generate images from pure noise by observing a wide range of examples. By simplifying the U-Net, we made the model more efficient without losing its ability to produce good images
To execute notebooks of this repository, you should install dependancies using :
pip install -r requirements.txt
We show here a simulation of batch of images that we generated using our model using MNIST dataset.
We trained two different models using the datasets MNIST and SVHN