Ver Anima V1
Trained base on Anima V0.1 base with LoRA_Easy_Training_Scripts. See about this version for training details.
Althrough Anima has improved its ability to genereate tattoo on the right position, still needs a lora to help it like penis tattoo or stomach tattoo. So I made a new version.
But the better way is to draw it with edit model LOL.
Ver NOOB V-PRED V3 EQ
Trained with improved parameters based on NOOB V-PRED EQ VAE exprimental version: https://civarchive.com/models/1858821/noobai-v-pred-10-with-eq-vae
In this version, I removed some trigger words because checkpoint already has the ability to generate them. Now it has 7 trigger words. See the examples first.
Ver NOOB V-PRED V2
Trained based on NOOB V-PRED 1.0 with OneTrainer. See about this version for training details. Thirteen trigger words set. Use color words to control the color of tattoo. See the examples first.
Pros
Better responsive to trigger words
Better artstyle
Concept enhancement
Cons
Many trials needed
Still has influence on art style
Ver NOOB V-PRED V1
Trained based on NOOB V-PRED 1.0. Thirteen trigger words set. Use color words to control the color of tattoo. See the examples first.
Introduction
A Pony based LyCORIS to enhance the concept Tattoo. Thirteen trigger words set. Use color words to control the color of tattoo. See the examples first.
Pros
Concept enhancement
Cons
Not responsive to trigger words, especially the penis tattoo, you need to upweight it or roll for it times to times
Many trials needed
Influence on art style
PS: Be sure to use underline when typing trigger words. The ones without underline (e.g. penis tattoo) will work but not effectively as the underline one (e.g. penis_tattoo)
Description
[[subsets]]
caption_extension = ".txt"
caption_tag_dropout_rate = 0.05
image_dir = "G:/dataset/concept/tattoo_enhancement/3_full-body_tattoo"
keep_tokens = 1
name = "3"
num_repeats = 3
random_crop_padding_percent = 0.05
shuffle_caption = true
[[subsets]]
caption_extension = ".txt"
caption_tag_dropout_rate = 0.05
image_dir = "G:/dataset/concept/tattoo_enhancement/5_breast_tattoo"
keep_tokens = 1
name = "5"
num_repeats = 5
random_crop_padding_percent = 0.05
shuffle_caption = true
[[subsets]]
caption_extension = ".txt"
caption_tag_dropout_rate = 0.05
image_dir = "G:/dataset/concept/tattoo_enhancement/5_leg_tattoo"
keep_tokens = 1
name = "5_1790613366622097100"
num_repeats = 5
random_crop_padding_percent = 0.05
shuffle_caption = true
[[subsets]]
caption_extension = ".txt"
caption_tag_dropout_rate = 0.05
image_dir = "G:/dataset/concept/tattoo_enhancement/5_penis_tattoo"
keep_tokens = 1
name = "5_1790613366636606800"
num_repeats = 5
random_crop_padding_percent = 0.05
shuffle_caption = true
[[subsets]]
caption_extension = ".txt"
caption_tag_dropout_rate = 0.05
image_dir = "G:/dataset/concept/tattoo_enhancement/5_side_tattoo"
keep_tokens = 1
name = "5_1790613366652106400"
num_repeats = 5
random_crop_padding_percent = 0.05
shuffle_caption = true
[[subsets]]
caption_extension = ".txt"
caption_tag_dropout_rate = 0.05
image_dir = "G:/dataset/concept/tattoo_enhancement/5_stomach_tattoo"
keep_tokens = 1
name = "5_1790613366666623200"
num_repeats = 5
random_crop_padding_percent = 0.05
shuffle_caption = true
[[subsets]]
caption_extension = ".txt"
caption_tag_dropout_rate = 0.05
image_dir = "G:/dataset/concept/tattoo_enhancement/5_testicles_tattoo"
keep_tokens = 1
name = "5_1790613366681338900"
num_repeats = 5
random_crop_padding_percent = 0.05
shuffle_caption = true
[train_mode]
train_mode = "lora"
[general_args.args]
persistent_data_loader_workers = true
vae_batch_size = 5
pretrained_model_name_or_path = ""
mixed_precision = "bf16"
gradient_checkpointing = true
gradient_accumulation_steps = 1
seed = 42
max_data_loader_n_workers = 1
max_token_length = 225
prior_loss_weight = 1.0
sdpa = true
max_train_epochs = 20
cache_latents = true
cache_latents_to_disk = true
[general_args.dataset_args]
resolution = 1024
batch_size = 5
[network_args.args]
network_dim = 32
network_alpha = 16.0
min_timestep = 0
max_timestep = 1000
network_train_unet_only = true
[optimizer_args.args]
optimizer_type = "ProdigyPlusScheduleFree"
lr_scheduler = "constant"
loss_type = "l2"
learning_rate = 1.0
unet_lr = 1.0
max_grad_norm = 1.0
[saving_args.args]
output_dir = "G:/LoRA_Easy_Training_Scripts/output"
output_name = "tattoo_enhancement-anima1.0-V1-loha-dim32conv16AR0.5-NewTag"
save_precision = "bf16"
save_model_as = "safetensors"
save_every_n_epochs = 1
save_toml = true
save_toml_location = "G:/LoRA_Easy_Training_Scripts/output"
[logging_args.args]
log_prefix_mode = "disabled"
run_name_mode = "default"
[anima_args.args]
pretrained_model_name_or_path = "G:/sd-webui-forge-neo/models/Stable-diffusion/ANIMA/anima_baseV10.safetensors"
qwen3 = "G:/sd-webui-forge-neo/models/text_encoder/qwen_3_06b_base.safetensors"
vae = "G:/sd-webui-forge-neo/models/VAE/qwen_image_vae.safetensors"
qwen3_max_token_length = 512
t5_max_token_length = 512
timestep_sampling = "sigmoid"
sigmoid_scale = 1.0
[edm_loss_args.args]
edm2_loss_weighting = false
[bucket_args.dataset_args]
enable_bucket = true
min_bucket_reso = 256
max_bucket_reso = 2048
bucket_reso_steps = 64
[network_args.args.network_args]
conv_dim = 16
conv_alpha = 8.0
algo = "loha"
train_llm_adapter = "false"
[optimizer_args.args.optimizer_args]
betas = "0.95,0.99"
beta3 = "None"
weight_decay = "0"
weight_decay_by_lr = "True"
d0 = "2e-5"
d_coef = "1"
d_limiter = "True"
schedulefree_c = "0"
eps = "None"
split_groups = "True"
split_groups_mean = "False"
factored = "True"
factored_fp32 = "True"
use_bias_correction = "True"
use_stableadamw = "True"
use_schedulefree = "True"
use_speed = "False"
stochastic_rounding = "True"
fused_back_pass = "False"
use_cautious = "False"
use_grams = "False"
use_adopt = "False"
use_orthograd = "False"
use_focus = "False"