denoising-diffusion-pytorch

denoising-diffusion-pytorch copied to clipboard

Sharing the checkpoint

9

Hi, thanks for your great implementation. Can you share the checkpoint trained on FFHQ dataset? Thanks.

gwang-kim

Hi, thanks for your excellent project ! I want to evaluate the model during training by adding the following code in the train.py script, which leads to program stucking. ```...

garychan22

KeyError: "lmdb dataset does not have key b'256-02098'"

2

on running the train.py script on my custom dataset, the code reflects a following error KeyError: "lmdb dataset does not have key b'256-02098'"

rudransh2004

Thanks for your great work! I wonder if this repo is able to train with hybrid loss ($\lambda L_{vlb} + L_{simple}$) proposed in improved DDPM？

wileewang

Can you please provide the training code for 128?

zhou431496

Suggestions for implementing on a non-image task

1

The application I have in mind doesn't use images, but has a notion of high-resolution, low-resolution, and random noise. Curious if you have thoughts on how to implement a denoising...

sgbaird

Have you tried to train a FFHQ 1Kx1K model?

yaseryacoob

Script/Instructions for producing samples?

1

I couldn't find the script to produce samples like the one that exist in the ReadMe. Does it exist in the repo? If not, would you mind committing the evaluation...

Skylion007

Hi Rosinality, great work! Would it be possible to post a full set of the dependencies for this repo? This would extend the question already posed in the issues [here](https://github.com/rosinality/denoising-diffusion-pytorch/issues/14#issue-1443101646)....

MHRosenberg

Hi Rosinality. I was going through the U-Net architecture you implemented for DDPM. I couldn't understand why you have added one extra resnet block in the upsampling path while not...

pimakshay