Diffusion Models

This work was part of a term paper / project for the UMC203 Introduction to AI and ML course at IISc.

Contributors

• Om Prakash Choudhary
• Himesh Kumar Anant
• Jithendra Rao Kasibhatla
• Keval Pithadiya

Models Trained

We trained Denoising Diffusion Probabilitic Models using the HuggingFace diffusers library for the clean image prediction and the noise prediction objectives on the below datasets and image sizes

• CIFAR10 (32x32)
• CelebA (32x32)
• CelebA (128x128)
• MNIST (32x32)

The trained models can be downloaded from this Google Drive link: https://drive.google.com/drive/folders/1HmLK4hkkI6-Fl6vaHyX05PBe8U2FWl5E?usp=drive_link

Alternative HuggingFace link: https://huggingface.co/Om2005Prakash/Diffusion_Pre_Trained/tree/main

Requirements

HuggingFace diffusers Library

pip install diffusers[training]

Python Version Used: 3.13.2