paul-gaugin-sdxl-lora-03

This is a LyCORIS adapter derived from stabilityai/stable-diffusion-xl-base-1.0.

The main validation prompt used during training was:

ggn_style, A nude woman sits on a patterned cloth with crossed legs. She holds a small object in her hand. A low table with fruit is in front of her. Pink flowers are in the background, with mountains and palm trees beyond. Text is in the bottom left corner.

Validation settings

CFG: 4.2
CFG Rescale: 0.0
Steps: 20
Sampler: None
Seed: 42
Resolution: 1024x1024

Note: The validation settings are not necessarily the same as the training settings.

You can find some example images in the following gallery:

Prompt
unconditional (blank prompt)

Negative Prompt
blurry, cropped, ugly

Prompt
your prompt to validate on

Negative Prompt
blurry, cropped, ugly

Prompt
another prompt to validate on

Negative Prompt
blurry, cropped, ugly

Prompt
ggn_style, A nude woman sits on a patterned cloth with crossed legs. She holds a small object in her hand. A low table with fruit is in front of her. Pink flowers are in the background, with mountains and palm trees beyond. Text is in the bottom left corner.

Negative Prompt
blurry, cropped, ugly

The text encoder was not trained. You may reuse the base model text encoder for inference.

Training settings

Training epochs: 0
Training steps: 400
Learning rate: 5e-05
Effective batch size: 4
- Micro-batch size: 4
- Gradient accumulation steps: 1
- Number of GPUs: 1
Prediction type: epsilon
Rescaled betas zero SNR: False
Optimizer: optimi-lionweight_decay=1e-3
Precision: Pure BF16
Quantised: Yes: int8-quanto
Xformers: Not used
LyCORIS Config:

{
    "algo": "lokr",
    "multiplier": 1.0,
    "linear_dim": 10000,
    "linear_alpha": 1,
    "factor": 16,
    "apply_preset": {
        "target_module": [
            "Attention",
            "FeedForward"
        ],
        "module_algo_map": {
            "Attention": {
                "factor": 16
            },
            "FeedForward": {
                "factor": 8
            }
        }
    }
}

Datasets

paul-gaugin-sdxl-512

Repeats: 10
Total number of images: 85
Total number of aspect buckets: 8
Resolution: 0.262144 megapixels
Cropped: False
Crop style: None
Crop aspect: None

paul-gaugin-sdxl-1024

Repeats: 10
Total number of images: 85
Total number of aspect buckets: 6
Resolution: 1.048576 megapixels
Cropped: False
Crop style: None
Crop aspect: None

paul-gaugin-sdxl-512-crop

Repeats: 10
Total number of images: 85
Total number of aspect buckets: 1
Resolution: 0.262144 megapixels
Cropped: True
Crop style: random
Crop aspect: square

paul-gaugin-sdxl-1024-crop

Repeats: 10
Total number of images: 85
Total number of aspect buckets: 1
Resolution: 1.048576 megapixels
Cropped: True
Crop style: random
Crop aspect: square

Inference

import torch
from diffusers import DiffusionPipeline
from lycoris import create_lycoris_from_weights

model_id = 'stabilityai/stable-diffusion-xl-base-1.0'
adapter_id = 'pytorch_lora_weights.safetensors' # you will have to download this manually
lora_scale = 1.0
wrapper, _ = create_lycoris_from_weights(lora_scale, adapter_id, pipeline.transformer)
wrapper.merge_to()

prompt = "ggn_style, A nude woman sits on a patterned cloth with crossed legs. She holds a small object in her hand. A low table with fruit is in front of her. Pink flowers are in the background, with mountains and palm trees beyond. Text is in the bottom left corner."
negative_prompt = 'blurry, cropped, ugly'
pipeline.to('cuda' if torch.cuda.is_available() else 'mps' if torch.backends.mps.is_available() else 'cpu')
image = pipeline(
    prompt=prompt,
    negative_prompt=negative_prompt,
    num_inference_steps=20,
    generator=torch.Generator(device='cuda' if torch.cuda.is_available() else 'mps' if torch.backends.mps.is_available() else 'cpu').manual_seed(1641421826),
    width=1024,
    height=1024,
    guidance_scale=4.2,
    guidance_rescale=0.0,
).images[0]
image.save("output.png", format="PNG")

davidrd123
/

paul-gaugin-sdxl-lora-03