metadata

license: mit
datasets:
  - unography/SAM-LLAVA-55k
pipeline_tag: image-to-image
tags:
  - control-lora-v2
  - stable-diffusion

Model Card for lavinal712/sd-control-lora-segmentation

Model Description

This is controlnet weight trained on runwayml/stable-diffusion-v1-5 with segmentaion.

Training

This model was trained using a Segmented dataset based on the SAM-LLaVA-Captions10M Dataset. Stable Diffusion v1.5 checkpoint was used as the base model for the controlnet.

SAM-LLAVA-55k

Training Method

Train on SAM-LLAVA-55k for 55000 steps with batch size of 4.