My first attempt at training a LoRA network. Still learning the ins and outs, but I think this one came out well. I really like the modularity they have, so I'll probably do them more frequently in the future once I have a good workflow for it. If you're using Automatic1111, I highly recommend this extension for working with LoRAs dynamically with different models: https://github.com/kohya-ss/sd-webui-additional-networks
I've tried this one with multiple models and networks and it seems to be pretty versatile, the one issue I've run into is that result may get pixelated when upscaling or doing img2img, and if that's the case try using a different upscaling algorithm.
1.転載モデルは学習・共有目的のみで使用し、著作権は原作者に帰属します
2.モデルの認証は公式チャンネルでご連絡ください。クリエイターの権利保護に努めています クリックして認証
