ℹ️What is this | 这是什么
这是一个风格LoCon / LoRA，可以画出具有繁复结构的类蒸汽朋克/赛博朋克/... 反正就是机械风格的角色，取决于你填什么提示词。风格来源于Dylan Kowalski。
This is a style LoCon / LoRA which generate highly detailed, complex and steampunk-alike character. The artstyle is from 3D artist Dylan Kowalski.
训练素材包含10张图片，分辨率512*768。训练基础模型为Chillout
Dataset contains 10 pictures with resolution at 512*768, trained on chillout.
预览图片生成使用Eris，它和这个LoCon 配合的很好。
Previews are generated with Eris, which works very well for this LoCon.
更新了一些图片示例 / Updated some images.
添加了Epoch-06，对画风影响较弱。可自行尝试 / Added Epoch-06, these have lesser effect on style, for those who wish to reduce the artifacts.
💦Known Issues | 已知问题
🫱在部分Checkpoint下手画的非常答辩，建议在负面提示词中加入hand。
Hands are really bad (at least on some checkpoints), avoid generating hands by adding "hand" into the negative prompts, or use "portrait, upper body, close up," in the positives.
⚖️权重高于0.5会导致画面迅速崩坏，LoRA加载过多可能导致最终图片模糊。建议最高0.5，少加载LoRA。这个插件有可能缓解。
Weight above 0.5 tends to destroy image. Loading too many LoRA will produce blurry images. This might help.
👦有时候会拒绝画男性，在负面中添加“1girl, female”可以解决这个问题。
Sometimes it refuses to generate male character, add "1girl, female" in the negative to solve this.
❓How to use | 使用方法
✅必须提示词 / Mandatory trigger：
🔸byDylanKowalski,<lora:dylanKowalski:0.5>,
✳️可以搭配以下提示词 / Works well with：
🔸steampunk（for steampunk aesthetic）
🔸cyberpunk（for cyberpunk aesthetic）
🔸intricate details
🔸ornaments
🔸prosthesis
🔸clockwork
✳️其他和材质相关的提示词 / OTHER PROMTS INDICATE MATERIALS, BE CREATIVE. (EXAMPLE)
🔸porcelain
🔸titanium
🔸gemstones
🔸holographic
🔸...
📝其他信息与参数:
🔸Model: Eris / Darelites Fantasy Mix
🔸Image Resolution / 分辨率: 512*768
🔸Sampler / 采样器: DPM++ 2M Karras,
🔸Sampling Steps / 采样步数: 20-40,
🔸Hires fix / 高分辨率修复算法: R-ESRGAN 4+.
🧡 希望你喜欢这个模型。I hope you enjoy it as much as I do. 🧡
Description
FAQ
Comments (5)
However, it seems that there are too many detailed elements and the result is that the image is not strong enough
Not sure what error is this. Fail to load.
LoRA weight_unet: 0.7, weight_tenc: 0.7, model: dylanKowalski6(c15a71db05c4)
dimension: 8, alpha: 8.0, multiplier_unet: 0.7, multiplier_tenc: 0.7
create LoRA for Text Encoder: 72 modules.
create LoRA for U-Net: 192 modules.
original forward/weights is backed up.
Error running process: E:\stable-diffusion-webui\extensions\sd-webui-additional-networks\scripts\additional_networks.py
Traceback (most recent call last):
File "E:\stable-diffusion-webui\modules\scripts.py", line 417, in process
script.process(p, *script_args)
File "E:\stable-diffusion-webui\extensions\sd-webui-additional-networks\scripts\additional_networks.py", line 206, in process
network, info = lora_compvis.create_network_and_apply_compvis(du_state_dict, weight_tenc, weight_unet, text_encoder, unet)
File "E:\stable-diffusion-webui\extensions\sd-webui-additional-networks\scripts\lora_compvis.py", line 102, in create_network_and_apply_compvis
state_dict = network.apply_lora_modules(du_state_dict) # some weights are applied to text encoder
File "E:\stable-diffusion-webui\extensions\sd-webui-additional-networks\scripts\lora_compvis.py", line 289, in apply_lora_modules
state_dict = LoRANetworkCompvis.convert_state_dict_name_to_compvis(self.v2, du_state_dict)
File "E:\stable-diffusion-webui\extensions\sd-webui-additional-networks\scripts\lora_compvis.py", line 193, in convert_state_dict_name_to_compvis
compvis_name = LoRANetworkCompvis.convert_diffusers_name_to_compvis(v2, tokens[0])
File "E:\stable-diffusion-webui\extensions\sd-webui-additional-networks\scripts\lora_compvis.py", line 182, in convert_diffusers_name_to_compvis
assert cv_name is not None, f"conversion failed: {du_name}. the model may not be trained by sd-scripts."
AssertionError: conversion failed: lora_unet_down_blocks_0_downsamplers_0_conv. the model may not be trained by sd-scripts.
Details
Files
dylanKowalski-000006.safetensors
Mirrors
ScifiProsthesisStyle_epoch000006.safetensors
ScifiProsthesisStyle_epoch000006.safetensors
ScifiProsthesisStyle_epoch000006.safetensors
dylanKowalski-000006.safetensors
styleLoconLoraScifiProsthesis_epoch000006.safetensors
ScifiProsthesisStyle_epoch000006.safetensors
ScifiProsthesisStyle_epoch000006.safetensors
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.