running training num train images * repeats: 44 num reg images: 0 num batches per epoch: 44 num epochs: 57 batch size per device: 1 gradient accumulation steps: 1 total optimization steps: 2500










Another attempt at a flux lora. It'll make fusion generators from fallout 4/76. As always if anyone has any tips on how to make this look better or any of my loras. Let me know. It's all guesswork right now.
This one works but it seems I likely didn't have enough realistic representations in my training data as these all come out a bit 3D render/video game style. Oh well
I have earlier epochs saved and they have less of a 3D style. Can upload if anyone wants.
This version:
Training Data:
22 unique images, repeated twice per epoch (44 image presentations per epoch).
No regularization images were used.
Training Setup:
Epochs: 57
Batch Size: 1
Gradient Accumulation Steps: 1
Total Optimization Steps: 2500
running training num train images * repeats: 44 num reg images: 0 num batches per epoch: 44 num epochs: 57 batch size per device: 1 gradient accumulation steps: 1 total optimization steps: 2500
1. 재게시된 모델의 권리는 원 제작자에게 있습니다.
2. 모델 원작자가 모델을 인증받으려면 공식 채널을 통해 SeaArt.AI 직원에게 문의하세요. 저희는 모든 창작자의 권리를 보호하기 위해 노력합니다. 인증하러 이동
