squares
discard-icon
vip-icon

Yurucamp_InuyamaAoi

v3.0
Yurucamp_InuyamaAoi_v1
Reprint
Last Update:2023-12-28
#Anime
#Character
#yurucamp
#girls

About

Character: Inuyama Aoi (Yurucamp)

Update History

2023-12-05

Add v3.
Detail: Read "About this version"

Quick Translation
Comment 0
Creations Number (0)
Favorite (0)
Download (0)
Model Details
Type
LORA
Rating
0
Publish Time
2024-01-10
Base Model
SD 1.5
Trigger Words
inuyama_aoi
Model Source: civitai
Version Introduction

inuko v3

prompts

inuyama_aoi, mediam_hair, breasts, bangs, brown_hair, hair_ornament,, side_ponytail, swept_bangs, thick_eyebrows, hair_bobbles, skin_fang, green eyes , fang, sailor_collar,, sweater, , school uniform, yellow sweater,  , serafuku, pleated skirt, checkered clothes, white_neckerchief

# sd -scripts command

Generate command : RedRayz/Kohya_lora_param_gui: GUI for kohya-ss sd-scripts (github.com)

  • LoRA-C3Lier (conv2d)

  • pagedAdamW8bit

  • Dim64 , Alpha32

  • CUT: IN07-08, OUT10-11

accelerate launch --num_cpu_threads_per_process 4 train_network.py --pretrained_model_name_or_path "C:\ProgramTools\__GenerativeAI\StableDiffusionModels\nai.ckpt" --train_data_dir "H:\Resources\Resources_AI\forLeaningImages\Yurucamp\inuyama_aoi\v2" --output_dir "H:\Resources\Resources_AI\__Models\__Lola\__MyLoRAs" --network_module "networks.lora" --network_args "conv_dim=64" "conv_alpha=32"  "down_lr_weight=1,1,1,1,1,1,1,0,0,1,1,1" "mid_lr_weight=1" "up_lr_weight=1,1,1,1,1,1,1,1,1,1,0,0" --xformers --persistent_data_loader_workers --cache_latents --max_token_length 225 --max_data_loader_n_workers 1 --enable_bucket --save_model_as "safetensors" --lr_scheduler_num_cycles 4 --mixed_precision "fp16" --learning_rate=1e-04 --text_encoder_lr=5e-05 --resolution 768 --train_batch_size 2 --max_train_epochs 12 --network_dim 64 --network_alpha 32 --shuffle_caption --keep_tokens 1 --save_every_n_epochs 1 --optimizer_type "PagedAdamW8bit" --lr_warmup_steps 200 --output_name "lora_yurucamp_inuyama_aoi_v3_pagedAdamW8bit_d64a32_NP5" --clip_skip 2 --save_precision "fp16" --lr_scheduler "cosine" --min_bucket_reso 320 --max_bucket_reso 1536 --caption_extension ".txt" --seed 42 --logging_dir "E:\Creative\Program\Github\__AI\Art\TraningLog\sd_scripts\log" --training_comment "lora_yurucamp_inuyama_aoi_v3_pagedAdamW8bit_d64a32_NP5" --sample_every_n_steps=537 --sample_prompts="H:\Resources\Resources_AI\forLeaningImages\__SamplePrompts\sample_prompts_basic1girl_inuko.txt" --sample_sampler=euler --log_prefix=inuko_v3

# Extract only the necessary layers

Tools: SuperMerger

enable only IN02_IN04_OUT03-06

  1. set to lbwpreset.txt

CHARA:0,0,1,1,0,0,0,0,1,1,1,1,0,0,0,0,0
  1. Enter text "LoRA_name:1:CHARA"

  2. Merge LoRA

  3. Done

# Additional information

I think you can get the same result even if you only learn IN02_IN04_OUT03-06 at the time of training.

View More
License Scope
Creative License Scope
Online Image Generation
Merge
Allow Downloads
Commercial License Scope
Sale or Commercial Use of Generated Images
Resale of Models or Their Sale After Merging