Instructions to use REPA-E/e2e-invae with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use REPA-E/e2e-invae with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline from diffusers.utils import load_image # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("REPA-E/e2e-invae", dtype=torch.bfloat16, device_map="cuda") prompt = "Turn this cat into a dog" input_image = load_image("https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/cat.png") image = pipe(image=input_image, prompt=prompt).images[0] - Notebooks
- Google Colab
- Kaggle
Where is the 4m checkpoint?
#2
by SwayStar123 - opened
The ImageNet sota was achieved with 800 epochs of training, which should be 4 million train steps with bs=256. Where can I find this checkpoint? I only see the 400k checkpoint here
It appears the run that resulted in SOTA FID froze the VAE and used this checkpoint.
SwayStar123 changed discussion status to closed