|
|
--- |
|
|
library_name: diffusers |
|
|
--- |
|
|
|
|
|
# Model Card for Model ID |
|
|
|
|
|
<!-- Provide a quick summary of what the model is/does. --> |
|
|
The model is able to generate 512x512 breast cancer pathology images based on generic text prompts, eg. "A histopathology image of breast cancer tissue." |
|
|
|
|
|
It is the fusion of Stable Diffusion 1.4 and a LORA fine-tuned on 28218 pathology images from the BRCA dataset. |
|
|
|
|
|
## Model Details |
|
|
|
|
|
### Model Description |
|
|
|
|
|
The model is trained on Stable Diffusion 1.4 with a UNet architecture and a pretrained VAE. It uses a batch size of 16 and lora rank of 32. |
|
|
|
|
|
### How to use |
|
|
|
|
|
```python |
|
|
pipe = DiffusionPipeline.from_pretrained( |
|
|
"RiddleHe/SD14_pathology_base", torch_dtype=torch.float16 |
|
|
) |
|
|
pipe.to('cuda') |
|
|
|
|
|
prompt = "A histopathology image of breast cancer tissue" |
|
|
``` |
|
|
|
