Network Module

LoRA

Use Base Model

WAI-illustrious - v16

Trigger words

splash_flowing_art

Image Processing Parameters

Repeat

1

Epoch

9

Save Every N Epochs

1

Training Parameters

Seed

-

Clip Skip

2

Text Encoder learning rate

0.00001

Unet/DiT learning rate

0.0001

LR Scheduler

constant_with_warmup

Optimizer

AdamW8bit

Network Dim

128

Network Alpha

64

Gradient Accumulation Steps

-

Label Parameters

Shuffle caption

true

Keep n tokens

1

Advanced Parameters

Noise offset

0.03

Multires noise discount

0.3

Multires noise iterations

10

conv_dim

256

conv_alpha

128

Batch Size

1

Steps: 4203 ·

Loss: 0.112 ·

Dataset number 467

Label model: wd-swinv2-tagger-v3