Training

settings

max pixel length of each image

1536

number of images

234

Steps:

3276

image tag generation model

wd-swinv2-tagger-v3

or

wd-v1-4-vit-tagger-v2

Parameter Settings

Network Module

DoRA

Use Base Model

WAI-illustrious - v16

Trigger words

splash_painting

Image Processing Parameters

Repeat

2

Epoch

7

Save Every N Epochs

1

Training Parameters

Seed

-

Clip Skip

2

Text Encoder learning rate

1

Unet/DiT learning rate

1

LR Scheduler

constant

Optimizer

Prodigy

Network Dim

64

Network Alpha

32

Gradient Accumulation Steps

-

Label Parameters

Shuffle caption

true

Keep n tokens

1

Advanced Parameters

Noise offset

0.03

Multires noise discount

0.1

Multires noise iterations

10

conv_dim

-

conv_alpha

-

Batch Size

1

Sample Image Settings

Prompt

splash_painting, 1girl, solo, long hair, breasts, looking at viewer, holding, hair between eyes, eyes, hair, ponytail, flower, weapon, horns, pants, official alternate costume, holding weapon, petals, mask, glowing, water, rain, wind, squatting, rope, glowing eyes, mouth mask, on one knee, glowing weapon, rope belt, sickle, akali, league_of_legends

Sampler

euler_ancestral