transformer_latent_diffusion icon indicating copy to clipboard operation
transformer_latent_diffusion copied to clipboard

Published 20 hours ago • verified publisher icon apapiu

Data source?

Open cloneofsimo opened this issue 1 year ago • 1 comments

Hi there! Im trying to make minRF, and there was a pointer to here, I was wondering what dataset you used for this! Thanks!

cloneofsimo avatar May 09 '24 16:05 cloneofsimo

Hey! Copying my answer from a previous issue - "The data - this is a big one - the full GRIT data might contain a lot of low quality images and/or prompts. Most of the data I used was either synthetic or filtered by CLIP aesthetic score. Try the mj_latents.npy and mj_text_emb.npy from here https://huggingface.co/apapiu/small_ldt/tree/main - this is higher quality synthetic data - I think about 600k examples if I remember correctly." Or you can use the data processing to download any data that has image and caption pairs from huggingface.

apapiu avatar May 09 '24 19:05 apapiu