Text-to-Image
Diffusers
Safetensors
English
StableDiffusionPipeline
diffusion
concept-erasure
stable-diffusion
esdu
golf_ball
Instructions to use DiffusionConceptErasure/esdu_golf_ball with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use DiffusionConceptErasure/esdu_golf_ball with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("DiffusionConceptErasure/esdu_golf_ball", dtype=torch.bfloat16, device_map="cuda") prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k" image = pipe(prompt).images[0] - Notebooks
- Google Colab
- Kaggle
- Local Apps
- Draw Things
- DiffusionBee
import torch
from diffusers import DiffusionPipeline
# switch to "mps" for apple devices
pipe = DiffusionPipeline.from_pretrained("DiffusionConceptErasure/esdu_golf_ball", dtype=torch.bfloat16, device_map="cuda")
prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k"
image = pipe(prompt).images[0]esdu_golf_ball
This is a concept-erased Stable Diffusion model using the Unconstrained Source Distillation (ESD-U) method to remove the concept "Golf Ball".
Method
Unconstrained Source Distillation (ESD-U) performs unconstrained distillation to remove concept information.
Usage
from diffusers import StableDiffusionPipeline
import torch
pipe = StableDiffusionPipeline.from_pretrained("ErasureResearch/esdu_golf_ball", torch_dtype=torch.float16).to("cuda")
prompt = "a photo of a golf_ball"
image = pipe(prompt).images[0]
image.save("erased_golf_ball.png")
Citation
If you use this model in your research, please cite:
@article{concept_erasure_2024,
title={Concept Erasure in Diffusion Models},
author={ErasureResearch Team},
journal={Proceedings of...},
year={2024}
}
- Downloads last month
- -