Nanakusa Nazuna - Yofukashi no uta [Pony, SDXL]版本v1.0 [Pony] (ID: 828850)

Nanakusa Nazuna - Yofukashi no uta

This SDXL/Pony character LoRA is trained on a collection of images of Nanakusa Nazuna from the series Yofukashi no uta.

DoRA [Pony] version is also available here.

See versions for recommended tags/triggers

Appearance:

pink_hair, short_hair, blue_eyes, ringed_eyes, twin_braids, small_breasts, black_nails

Default Outfit:

pink_hair, short_hair, blue_eyes, ringed_eyes, twin_braids, small_breasts, black_shorts, clothing_cutout, bare_shoulders, crop_top, black_thighhighs, black_nails

Maid Outfit:

pink_hair, short_hair, blue_eyes, ringed_eyes, twin_braids, small_breasts, black_shorts, maid_headdress

Inspired by these SD 1.5 LoRA's:

• https://civitai.com/models/25339/nanakusa-nazuna-or-yofukashi-no-uta

• https://civitai.com/models/297306/nazuna-nanakusa-yofukashi-no-uta

License

Fair AI Public License 1.0-SD

描述:

Trained on a different dataset to the SDXL v1 version.

This is part of a small training comparison test of different networks (LoRA, DoRA, LoKR), can reference this post for their comparisons - I can upload the LoKR version on request but I felt the LoRA/DoRA versions were better.

Kohya training config (toml):

bucket_no_upscale = true

bucket_reso_steps = 256

cache_latents = true

cache_latents_to_disk = true

caption_extension = ".txt"

clip_skip = 2

dynamo_backend = "no"

enable_bucket = true

epoch = 20

full_bf16 = true

gradient_accumulation_steps = 1

gradient_checkpointing = true

huber_c = 0.1

huber_schedule = "snr"

learning_rate = 1.0

logging_dir = "E:/work/LoRa_work/logging"

loss_type = "l2"

lr_scheduler = "cosine"

lr_scheduler_args = []

lr_scheduler_num_cycles = 1

lr_scheduler_power = 1

max_bucket_reso = 2048

max_data_loader_n_workers = 0

max_grad_norm = 1

max_timestep = 1000

max_token_length = 225

max_train_epochs = 20

max_train_steps = 153

metadata_author = "motimalu"

min_bucket_reso = 512

min_snr_gamma = 5

mixed_precision = "bf16"

multires_noise_discount = 0.3

multires_noise_iterations = 6

network_alpha = 32

network_args = []

network_dim = 32

network_module = "networks.lora"

no_half_vae = true

noise_offset_type = "Multires"

optimizer_args = []

optimizer_type = "Prodigy"

output_dir = "D:/model_output"

output_name = "pony_nanakusa_nazuna_lora_v1"

pretrained_model_name_or_path =

"C:/workspace/webui_forge_cu121_torch21/webui/models/Stable-diffusion/ponyDiffusionV6XL_v6Start

WithThisOne.safetensors"

prior_loss_weight = 1

resolution = "1024,1024"

sample_every_n_epochs = 10

sample_prompts = "D:/model_output\\prompt.txt"

sample_sampler = "euler_a"

save_every_n_epochs = 20

save_model_as = "safetensors"

save_precision = "bf16"

seed = 339491249

text_encoder_lr = 1

train_batch_size = 8

train_data_dir = "C:/data/nanakusa_nazuna"

unet_lr = 1

xformers = true

训练词语: nanakusa nazuna \(yofukashi no uta\),blue eyes, ringed eyes, vampire, braided hair rings, blunt bangs, short hair,sleeveless turtleneck, hood, black cloak, crop top, cleavage cutout, jacket, short shorts, black shorts, thigh boots, black thighhighs