YAML Metadata
Warning:
empty or missing yaml metadata in repo card
(https://huggingface.co/docs/hub/model-cards#model-card-metadata)
ESD Model - Concept Erasing
This is an ESD (Erasing Concepts from Diffusion Models) model trained to erase specific concepts.
Model Details
Erased Concepts
- erase_concept: hate, harassment, violence, suffering, humiliation, harm, suicide, sexual, nudity, bodily fluids, blood, obscene gestures, illegal activity, drug use, theft, vandalism, weapons, child abuse, brutality, cruelty
- erase_concept_from: hate, harassment, violence, suffering, humiliation, harm, suicide, sexual, nudity, bodily fluids, blood, obscene gestures, illegal activity, drug use, theft, vandalism, weapons, child abuse, brutality, cruelty
- train_method: esd-u
Usage
This model can be used with the ESD technique to erase concepts from Stable Diffusion.
# Example code to use this model
from diffusers import StableDiffusionPipeline
from safetensors.torch import load_file
import torch
# Load the base model
pipe = StableDiffusionPipeline.from_pretrained("CompVis/stable-diffusion-v1-4", torch_dtype=torch.float16).to("cuda")
# Load the ESD parameters
esd_params = load_file("path/to/downloaded/model.safetensors")
# Apply ESD parameters to model
# (implementation depends on your ESD setup)
Inference Providers
NEW
This model isn't deployed by any Inference Provider.
๐
Ask for provider support