AingDiffusion XL v1.4 uses the Prodigy optimizer to train the model, this making sure that the model learns everything it needs to know by adjusting its learning rate.
Finetuning Layer Details
- Dataset count: 41,671 images
- Masterpiece-rated: 5,814
- Best Quality-rated: 3,155
- Normal Quality-rated: 5,353
- Low Quality-rated: 5,790
- Worst Quality-rated: 21,559
- GPU: 1 x RTX 4090 (36 hours of training)
- Batch size: 12
- Gradient Accumulation: 3
- Equivalent batch size: 36
- Epoch: 10
- Used epoch: 8
- LyCORIS Network: LoKr
- Network Dim: 32
- Network Alpha: 16
- Conv Dim: 16
- Conv Alpha: 8
NOTE: Dim settings used is to trigger full matrix mode on LoKr, not actually using mentioned Dim and Alpha.
- Optimizer: Prodigy
- LR Scheduler: Cosine annealing warmup restarts
- Restarts: 10
- Min. LR: 1e-06
- Learning Rate: 1
- Min SNR gamma: 5
- Other optimizer arguments:
- Betas: 0.9, 0.99
- Weight_decay: 0.01
- decouple: True
- use_bias_correction: True
- safeguard_warmup: True
- d_coef: 1
This LyCORIS was then merged with AingBase on 80% weight.
Quality tags ranking system is inspired by CagliostroLab’s scoring system.
