安装 https://github.com/toshiaki1729/stable-diffusion-webui-dataset-tag-editor.git 数据集标签编辑器以批量处理图片
Install https://github.com/toshiaki1729/stable-diffusion-webui-dataset-tag-editor.git to batch process images
使用tagger获得更好的tag描述
Use tagger to get better tag descriptions
缩小训练的图片以提高barch大小，对大数据集的收敛效果更好
Shrink the training images to increase the barch size for better convergence on large data sets
红线在1024*576 bs=4情况下训练 黄线在320*192 bs=50情况下训练
数据集的采集可能涉及侵权，该模型只用于学习交流，请于下载24小时内后删除
The collection of the data set may involve infringement, the model is only for learning and exchange, please delete after 24 hours of download
Description
FAQ
Comments (5)
Hi, I'm confused. This is a lora but why the description is about training images? Is it mandatory to install the .git from the description before this lora can be used?
TomLucidorDec 4, 2023
It is a sample result LoRA files for seeing if a LoRA can capture multiple art styles from the designated artists
请问你用的是哪个大模型来训练的
谢谢