Simpletuner Lora

🚀 simpletuner-lora

simpletuner-lora 是一個基於 black-forest-labs/FLUX.1-schnell 的 LyCORIS 適配器，可用於文本到圖像、圖像到圖像等任務。

🚀 快速開始

本項目是一個從 black-forest-labs/FLUX.1-schnell 派生而來的 LyCORIS 適配器。

訓練期間使用的主要驗證提示為：

A photo-realistic image of woozi_seventeen

✨ 主要特性

• 基於 black-forest-labs/FLUX.1-schnell 模型派生，可用於文本到圖像等任務。
• 提供了訓練和驗證的詳細設置。
• 包含推理代碼示例。

📦 安裝指南

文檔未提供具體安裝步驟，暫不展示。

💻 使用示例

基礎用法

import torch
from diffusers import DiffusionPipeline
from lycoris import create_lycoris_from_weights


def download_adapter(repo_id: str):
    import os
    from huggingface_hub import hf_hub_download
    adapter_filename = "pytorch_lora_weights.safetensors"
    cache_dir = os.environ.get('HF_PATH', os.path.expanduser('~/.cache/huggingface/hub/models'))
    cleaned_adapter_path = repo_id.replace("/", "_").replace("\\", "_").replace(":", "_")
    path_to_adapter = os.path.join(cache_dir, cleaned_adapter_path)
    path_to_adapter_file = os.path.join(path_to_adapter, adapter_filename)
    os.makedirs(path_to_adapter, exist_ok=True)
    hf_hub_download(
        repo_id=repo_id, filename=adapter_filename, local_dir=path_to_adapter
    )

    return path_to_adapter_file
    
model_id = 'black-forest-labs/FLUX.1-schnell'
adapter_repo_id = 'MyKPopWishList/simpletuner-lora'
adapter_filename = 'pytorch_lora_weights.safetensors'
adapter_file_path = download_adapter(repo_id=adapter_repo_id)
pipeline = DiffusionPipeline.from_pretrained(model_id, torch_dtype=torch.bfloat16) # loading directly in bf16
lora_scale = 1.0
wrapper, _ = create_lycoris_from_weights(lora_scale, adapter_file_path, pipeline.transformer)
wrapper.merge_to()

prompt = "A photo-realistic image of woozi_seventeen"


## Optional: quantise the model to save on vram.
## Note: The model was not quantised during training, so it is not necessary to quantise it during inference time.
#from optimum.quanto import quantize, freeze, qint8
#quantize(pipeline.transformer, weights=qint8)
#freeze(pipeline.transformer)
    
pipeline.to('cuda' if torch.cuda.is_available() else 'mps' if torch.backends.mps.is_available() else 'cpu') # the pipeline is already in its target precision level
model_output = pipeline(
    prompt=prompt,
    num_inference_steps=4,
    generator=torch.Generator(device='cuda' if torch.cuda.is_available() else 'mps' if torch.backends.mps.is_available() else 'cpu').manual_seed(42),
    width=1024,
    height=1024,
    guidance_scale=3.0,
).images[0]

model_output.save("output.png", format="PNG")

高級用法

文檔未提供高級用法示例，暫不展示。

📚 詳細文檔

驗證設置

• CFG: 3.0
• CFG Rescale: 0.0
• Steps: 4
• Sampler: FlowMatchEulerDiscreteScheduler
• Seed: 42
• Resolution: 1024x1024
• Skip-layer guidance:

注意：驗證設置不一定與訓練設置相同。

你可以在以下圖庫中找到一些示例圖像：

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