The key to this might be using a high repeat number and low learning rate, as well as a large enough LoRA capacity (I used 32/16). Most importantly, it rarely distorts her eyes. This version produces images extremely close to the anime art style. For v2, include bangs, hair between eyes as well (the model will generate her wearing glasses without bangs if not prompted) To generate school uniform: student uniform, gray dress, red ribbon, neck ribbon, pinafore dress, white shirt This is important, as previous models I trained for Kanon as well as other characters lacked the ability to generate sideway images. 20-30 images are chosen from sideway angles (booru tag from side). Trained using around 150 anime screenshots from season 1 and season 2. The final checkpoint is created by averaging the checkpoints of epoch 12 and 8: 0.8*epoch_12 + 0.2*epoch_8.
Pretrained_model_name_or_path = "/content/ashleymix-v2.safetensors" Logging_dir = "/content/drive/MyDrive/lora_training/log" Output_dir = "/content/drive/MyDrive/lora_training/output/shibukano" The model is trained with below parameters and repeat 8:
Inference should work on any model, but Meina_Hentai might give the best result (it is a good SFW model, don't get fooled by the name). The difference is that it is fine-tuned from my custom checkpoint with 0.7*Meina_Hentai + 0.3*AnyLoRA. This new version is trained on the same dataset as before, with 151 anime screencaps.
0 kommentar(er)
