Skip to content
GitLab
Explore
Sign in
Primary navigation
Search or go to…
Project
Unconditional Diffusion
Manage
Activity
Members
Code
Repository
Branches
Commits
Tags
Repository graph
Compare revisions
Locked files
Deploy
Model registry
Analyze
Model experiments
Help
Help
Support
GitLab documentation
Compare GitLab plans
Community forum
Contribute to GitLab
Provide feedback
Terms and privacy
Keyboard shortcuts
?
Snippets
Groups
Projects
Show more breadcrumbs
Diffusion Project
Unconditional Diffusion
Repository graph
Repository graph
You can move around the graph by using the arrow keys.
main
Select Git revision
Branches
3
celebAHQ
ddpm-diffusers
main
default
protected
3 results
Begin with the selected commit
Created with Raphaël 2.2.0
13
Aug
10
8
31
Jul
30
26
23
22
21
19
18
10
4
30
Jun
29
27
26
25
24
23
21
20
19
15
14
12
11
10
8
2
1
24
May
21
16
15
14
13
11
7
27
Apr
Add missing LHQ README samples
main
main
add UMAP for CelebAHQ
update UMAP for LHQ
update UMAP for LHQ ResNet
update UMAP for LHQ CLIP
minor changes to possible bugs in dataloader and training function
add UMAP viz on LHQ ResNet features
add UMAP viz on LHQ CLIP features
delete redundant UNet models
solved minor bugs, added images to README
eval update: eval directory, main, experiment_creator, readme
sampling function with image reconstruction
update kNN, add psnr, ssim metric
forgot to update no transform for the celebAHQ dataloader
add .ipynb_checkpoints to .gitignore
Add the CelebAHQ dataloader to main branch
update flexibility with diff input shape, output directories, readme
updated README
add full evaluation pipeline
insert original dataloader for training on Landscapes (LHQ256).
unnecessary merge conflict all_unets.py
Unconditional Diffusion Model used to train on CelebA HQ (Celebrity Faces)
celebAHQ
celebAHQ
add CelebAHQ dataloader
corrected model
updated some comments
added n times sampling possibility
Merge branch 'main' of https://git.rwth-aachen.de/diffusion-project/diffusion_project
Implemented evaluation function, now computes FID, IS and KID for a given sample of generated images w.r.t. testing data. Fixed device porblems with cosine noise scheduler
fix sample flag
small update to README.md
sample from all checkpoints at once
save raw tensor, normalize images,histogram
storing imagetensor and normalized images
Added save states for the EMA model when training from checkpoint
minor bug fixes
Merge branch 'ddpm-diffusers'
further augmentation added
ddpm-diffusers
ddpm-diffusers
fully functional unet + implementation in pipeline
EMA training added. Fixed some minor buggs.
Merge remote-tracking branch 'origin/ddpm-diffusers' into ddpm-diffusers
Loading