Stable diffusion fine-tune repo for Paris shopfront experiment (This Place Does Exist)
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
 
 
 
Ruben van de Ven a26269d7d1
Add notebook with comments for Lambda deployment
3 months ago
assets Finetuning (#11) 5 months ago
configs Additions to fine-tune TPDE dataset 3 months ago
data coco meta 8 months ago
examples update prior notebook 5 months ago
ldm Update modules.py 4 months ago
models add configs for training unconditional/class-conditional ldms 1 year ago
scripts Update slimify.py 4 months ago
.gitignore Finetuning (#11) 5 months ago
LICENSE add LICENSE 1 year ago
README.md Add notebook with comments for Lambda deployment 3 months ago
main.py Additions to fine-tune TPDE dataset 3 months ago
notebook_helpers.py add code 1 year ago
paris_dataloader.py Additions to fine-tune TPDE dataset 3 months ago
requirements.txt Finetuning (#11) 5 months ago
setup.py add code 1 year ago
tpde_finetune.ipynb Add notebook with comments for Lambda deployment 3 months ago

README.md

Experiments with Stable Diffusion

Fine-tune stable diffusion with images from Paris. Part of the This Place Does Exist experiment.

This repository extends and adds to the fine-tune repo by Justin Pinkney which in turn extends the original training repo for Stable Diffusion.

Currently it adds:

Fine tuning

Makes it easy to fine tune Stable Diffusion on your own dataset. For example generating new Pokemon from text:

Girl with a pearl earring, Cute Obama creature, Donald Trump, Boris Johnson, Totoro, Hello Kitty

For a step by step guide see the notebook in this repo (adapted from Lambda Labs examples repo.)