tanelp/tiny-diffusion
A minimal PyTorch implementation of probabilistic diffusion models for 2D datasets.
Jupyter Notebook
Issues
- 1
About reconstructing from noise?
#5 opened by Sreerag-ibtl - 0
appreciation for this great mini implementation of diffusion and some question
#8 opened by ShellCodingGH - 3
What is the reason for calculating the loss and pass it to the backward function as gradient parameter?
#3 opened by Eliyas0007 - 1
The current model configuration doesn't work well on the line dataset because the timestep is not large enough.
#4 opened by bopan3 - 0
- 0
amazing work
#1 opened by nile649