Outer-Spatial/nathan2

1 2---3base_model: stabilityai/stable-diffusion-xl-base-1.04instance_prompt: abbeyy5tags:6- stable-diffusion-xl7- stable-diffusion-xl-diffusers8- text-to-image9- diffusers10- lora11inference: false12datasets:13- Outer-Spatial/nathan214---15    16# LoRA DreamBooth - Outer-Spatial/nathan217These are LoRA adaption weights for stabilityai/stable-diffusion-xl-base-1.0 trained on @fffiloni's SD-XL trainer. 18The weights were trained on the concept prompt: 19```20abbeyy21```  22Use this keyword to trigger your custom model in your prompts. 23LoRA for the text encoder was enabled: False.24Special VAE used for training: madebyollin/sdxl-vae-fp16-fix.25## Usage26Make sure to upgrade diffusers to >= 0.19.0:27```28pip install diffusers --upgrade29```30In addition make sure to install transformers, safetensors, accelerate as well as the invisible watermark:31```32pip install invisible_watermark transformers accelerate safetensors33```34To just use the base model, you can run:35```python36import torch37from diffusers import DiffusionPipeline, AutoencoderKL38device = "cuda" if torch.cuda.is_available() else "cpu"39vae = AutoencoderKL.from_pretrained('madebyollin/sdxl-vae-fp16-fix', torch_dtype=torch.float16)40pipe = DiffusionPipeline.from_pretrained(41    "stabilityai/stable-diffusion-xl-base-1.0",42    vae=vae, torch_dtype=torch.float16, variant="fp16",43    use_safetensors=True44)45pipe.to(device)46# This is where you load your trained weights47specific_safetensors = "pytorch_lora_weights.safetensors"48lora_scale = 0.949pipe.load_lora_weights(50    'Outer-Spatial/nathan2', 51    weight_name = specific_safetensors,52    # use_auth_token = True 53)54prompt = "A majestic abbeyy jumping from a big stone at night"55image = pipe(56    prompt=prompt, 57    num_inference_steps=50,58    cross_attention_kwargs={"scale": lora_scale}59).images[0]60```61