2.0版本
使用了1024*1024的尺寸进行训练，试图解决“克隆人”的问题
使用了400张单人的图片训练训练
优化的模型大小，仅有27M
触发词：1girl,kamisato_ayaka,solo,tassel,breastplate,japanese_armor,skirt,japanese_clothes,dress,armored_dress,
2.0 Version:
Trained using a resolution of 1024x1024 in an attempt to address the issue of "clone images."
Trained on a dataset of 400 individual images.
Optimized model size, only 27MB.
Trigger words: 1girl, kamisato_ayaka, solo, tassel, breastplate, japanese_armor, skirt, japanese_clothes, dress, armored_dress.
1.5版本：
本模型版本采取高质量的角色训练集训练，比上一个版本效果提升了20%
对于官方服装可能没有很好的准确度，因为在训练集中官方服装的占比相当少
神里绫华太难训练了...
1.5 Version:
This model version was trained on a high-quality character dataset, resulting in a 20% improvement compared to the previous version.
It may not have high accuracy for official costumes because the proportion of official costumes in the training dataset is relatively low.
训练模型：SDXL1.0官模
测试模型：Kohaku-XL
建议权重：0.6~0.8
Training model: SDXL1.0 Official Model
Testing model: Kohaku-XL
Recommended weights: 0.6~0.8
Description
FAQ
Comments (4)
An AI enthusiast。Welcome to join my Discord to discuss SDXL and model training issues
Discord : https://discord.com/invite/CFpyHceF52
ai交流QQ群：914818692
Hi. When I use this lora the pictures are drawn but I get a lot of errors in the console like this (used 9527 Detail Realistic XL 3.0):
...
lora key not loaded lora_unet_up_blocks_3_attentions_2_transformer_blocks_0_attn2_to_v.lora_up.weight
lora key not loaded lora_unet_up_blocks_3_attentions_2_transformer_blocks_0_ff_net_0_proj.alpha
lora key not loaded lora_unet_up_blocks_3_attentions_2_transformer_blocks_0_ff_net_0_proj.lora_down.weight
...
ERROR diffusion_model.middle_block.1.transformer_blocks.0.attn2.to_k.weight shape '[1280, 2048]' is invalid for input of size 983040
ERROR diffusion_model.middle_block.1.transformer_blocks.0.attn2.to_v.weight shape '[1280, 2048]' is invalid for input of size 983040
ERROR diffusion_model.output_blocks.3.1.proj_in.weight shape '[640, 640]' is invalid for input of size 1638400
ERROR diffusion_model.output_blocks.3.1.proj_out.weight shape '[640, 640]' is invalid for input of size 1638400
ERROR diffusion_model.output_blocks.3.1.transformer_blocks.0.attn2.to_k.weight shape '[640, 2048]' is invalid for input of size 983040
It's an error that I haven't encountered before, and I am thoroughly researching it. Are all your UI versions up to date? Do you experience any anomalies when using other XL Lora models
@Chenkin Other xl loras work with no errors. Actually this is the first time I see such an error. ComfyUI is not the newest, but less than a month old.
Details
Available On (3 platforms)
Same model published on other platforms. May have additional downloads or version variants.