RiddleHe/SD14_pathology_lora

<!-- This model card has been generated automatically according to the information the training script had access to. You should probably proofread and complete it, then remove this comment. -->

LoRA text2image fine-tuning - RiddleHe/SD14pathologylora

These are LoRA adaption weights for CompVis/stable-diffusion-v1-4. The weights were fine-tuned on the None dataset. You can find some example images in the following.

<table> <tr> <td><img src="./image1.png"></td> <td><img src="./image2.png"></td> <td><img src="./image_3.png"></td> </tr> </table>

Intended uses & limitations

How to use

pipe = DiffusionPipeline.from_pretrained(
  "CompVis/stable-diffusion-v1-4", torch_dtype=torch.float16
)

pipe.load_lora_weights("RiddleHe/SD14_pathology_lora")
pipe.to('cuda')

prompt = "A histopathology image of breast cancer tissue"

Limitations and bias

[TODO: provide examples of latent issues and potential remediations]

Training details

This model is trained on 28216 breast cancer tissue images from the BRCA dataset.