ModelsLab/UrangDiffusion V3.1 - V1.1

Training Configurations

Pretraining:

- Dataset size: ~35,000 images

- GPU: 1xA100

- Optimizer: AdaFactor

- Unet Learning Rate: 2.5e-6

- Text Encoder Learning Rate: 1.25e-6

- Batch Size: 48

- Gradient Accumulation: 1

- Epoch: 10

Finetuning:

- Dataset size: ~3,100 images

- GPU: 1xA100

- Optimizer: AdaFactor

- Unet Learning Rate: 2e-6

- Text Encoder Learning Rate: - (Train TE set to False)

- Batch Size: 48

- Gradient Accumulation: 1

- Epoch: 10

- Noise Offset: 0.0357

Added Series

Wuthering Waves, Zenless Zone Zero and hololiveEN -Justice-