transformer_latent_diffusion icon indicating copy to clipboard operation
transformer_latent_diffusion copied to clipboard

Can you share full training code? Thank you!

Open ericwudocomoi opened this issue 2 years ago • 3 comments

Hi,

Thanks for the great work. I tried to run the train.py but "val_encodings.npy" does seems to be generated by data.py. Can you share the complete data generation/training code, thank you!

ericwudocomoi avatar Jan 20 '24 04:01 ericwudocomoi

Hi @ericwudocomoi I added a notebook in the https://github.com/apapiu/transformer_latent_diffusion?tab=readme-ov-file#usage tab: if you look in the notebook it will download some already preprocessed data including the val encodings and do a train run. I will shortly add a notebook to process a random dataset as well.

apapiu avatar Jan 21 '24 16:01 apapiu

Thank you!

ericwudocomoi avatar Jan 21 '24 17:01 ericwudocomoi

@ericwudocomoi Ok added another notebook in the https://github.com/apapiu/transformer_latent_diffusion?tab=readme-ov-file#usage subsection that should help you preprocess the images and text on your own dataset. Let me know if this helps and you're able to run it on your own.

apapiu avatar Jan 21 '24 18:01 apapiu