Instructions to use ZihengWang/ControlGlaucoma with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use ZihengWang/ControlGlaucoma with Diffusers:
pip install -U diffusers transformers accelerate
from diffusers import ControlNetModel, StableDiffusionControlNetPipeline controlnet = ControlNetModel.from_pretrained("ZihengWang/ControlGlaucoma") pipe = StableDiffusionControlNetPipeline.from_pretrained( "fill-in-base-model", controlnet=controlnet ) - Notebooks
- Google Colab
- Kaggle
| license: apache-2.0 | |
| tags: | |
| - controlnet | |
| - stable-diffusion | |
| - fundus | |
| - glaucoma | |
| - medical-imaging | |
| # ControlGlaucoma — Pre-trained Weights | |
| Pre-trained weights for [ControlGlaucoma](https://github.com/WANG-ZIHENG/ControlGlaucoma) — vCDR-controllable ControlNet on Stable Diffusion 2.1 for SLO fundus generation. | |
| | Folder | Description | | |
| |---|---| | |
| | `controlnet_vpred/` | Main weight. ControlNet on SD 2.1 (v-prediction). | | |
| | `controlnet_eps/` | ε-prediction variant (compatible with SD 2.1-base, 512 px). | | |
| | `segman_b/` | SegMAN-B segmentation model used as the reward signal during training. | | |
| ```bash | |
| huggingface-cli download ZihengWang/ControlGlaucoma \ | |
| --local-dir ./checkpoints \ | |
| --local-dir-use-symlinks False | |
| ``` | |
| See the [GitHub repository](https://github.com/WANG-ZIHENG/ControlGlaucoma) for code and training / inference commands. | |
| ## License | |
| Apache-2.0. | |