Updated
Used PonyXL to train this one. Updated dataset to achieve better performance. It enhanced this concept on PonyXL and related models. See the examples first.
I should say the Pony model fixed many problems I met. I do advice people to give it a try.
Introduction
An experimental LyCORIS to generate futanari filling condom. Use the key words to control the size of penis and condom.
Still a lot of room for improvement.
Pros
Can be combined with multiple Loras
Little impact on the art style
Cons
Using 3D materials for training, the art style may become strange under high weights
Twisted penises and multiple penises
Bad glan shape
Lots of trials needed
How to use it
I recommand using it in weight 1-1.2.
Training details
Traingset is of about 107 images. Half of them are chosen as regularization images. Total steps are about 7600.
resolution=768*768
batch_size=2
epoch=10
network_dim=32
network_alpha=16
conv_dim=8
conv_alpha=4
clip_skip=2
Using Prodigy as optimizer, setting as the author recommend.
Description
Used PonyXL to train this one. Updated dataset to achieve better performance. It enhanced this concept on PonyXL and related models. See the examples first.
I should say the Pony model fixed many problems I met. I do advice people to give it a try.
Parameters:
{ "LoRA_type": "LyCORIS/LoCon", "LyCORIS_preset": "full", "adaptive_noise_scale": 0, "additional_parameters": "", "block_alphas": "", "block_dims": "", "block_lr_zero_threshold": "", "bucket_no_upscale": true, "bucket_reso_steps": 128, "bypass_mode": false, "cache_latents": false, "cache_latents_to_disk": false, "caption_dropout_every_n_epochs": 0.0, "caption_dropout_rate": 0, "caption_extension": ".txt", "clip_skip": 2, "color_aug": false, "constrain": 0.0, "conv_alpha": 4, "conv_block_alphas": "", "conv_block_dims": "", "conv_dim": 8, "dataset_config": "", "debiased_estimation_loss": false, "decompose_both": false, "dim_from_weights": false, "dora_wd": false, "down_lr_weight": "", "enable_bucket": true, "epoch": 10, "factor": -1, "flip_aug": false, "fp8_base": false, "full_bf16": false, "full_fp16": false, "gpu_ids": "", "gradient_accumulation_steps": "1", "gradient_checkpointing": true, "keep_tokens": 1, "learning_rate": 1.0, "log_tracker_config": "", "log_tracker_name": "", "logging_dir": "/root/autodl-tmp/output", "lora_network_weights": "", "lr_scheduler": "cosine", "lr_scheduler_args": "", "lr_scheduler_num_cycles": "", "lr_scheduler_power": "", "lr_warmup": 0, "max_bucket_reso": 4096, "max_data_loader_n_workers": "0", "max_grad_norm": 1, "max_resolution": "1024,1024", "max_timestep": 1000, "max_token_length": "75", "max_train_epochs": "", "max_train_steps": "", "mem_eff_attn": false, "mid_lr_weight": "", "min_bucket_reso": 1024, "min_snr_gamma": 0, "min_timestep": 0, "mixed_precision": "bf16", "model_list": "custom", "module_dropout": 0, "multi_gpu": false, "multires_noise_discount": 0.1, "multires_noise_iterations": 6, "network_alpha": 8, "network_dim": 16, "network_dropout": 0.1, "noise_offset": 0, "noise_offset_type": "Multires", "num_cpu_threads_per_process": 2, "num_machines": 1, "num_processes": 1, "optimizer": "Prodigy", "optimizer_args": "weight_decay=0.01 d_coef=1 use_bias_correction=True safeguard_warmup=False betas=0.9,0.99 decouple=True ", "output_dir": "/root/autodl-tmp/output", "output_name": "condom_filling-dropout", "persistent_data_loader_workers": false, "pretrained_model_name_or_path": "/root/kohya_ss/models/ponyDiffusionV6XL.safetensors", "prior_loss_weight": 1.0, "random_crop": false, "rank_dropout": 0, "rank_dropout_scale": false, "reg_data_dir": "", "rescaled": false, "resume": "", "sample_every_n_epochs": 1, "sample_every_n_steps": 0, "sample_prompts": "score_9, score_8_up, score_7_up, anime BREAK, hyper_coneja, 1girl, solo, long hair, breasts, open mouth, black hair, thighhighs, white background, navel, nipples, purple eyes, nude, penis, dark skin, spread legs, armpits, huge breasts, arms up, white thighhighs, dark-skinned female, uncensored, erection, squatting, testicles, arms behind head, condom, veins, futanari, veiny penis, large penis, used condom, huge penis, condom on penis, large testicles, horse penis --n source_pony, source_furry, source_cartoon, greyscale, from below --w 1024 --h 1024\n ", "sample_sampler": "euler_a", "save_every_n_epochs": 1, "save_every_n_steps": 0, "save_last_n_steps": 0, "save_last_n_steps_state": 0, "save_model_as": "safetensors", "save_precision": "bf16", "save_state": false, "scale_v_pred_loss_like_noise_pred": false, "scale_weight_norms": 0, "sdxl": true, "sdxl_cache_text_encoder_outputs": false, "sdxl_no_half_vae": true, "seed": "31337", "shuffle_caption": true, "stop_text_encoder_training_pct": 0, "text_encoder_lr": 1.0, "train_batch_size": 5, "train_data_dir": "/root/autodl-tmp/condom_filling", "train_norm": false, "train_on_input": true, "training_comment": "", "unet_lr": 1.0, "unit": 1, "up_lr_weight": "", "use_cp": false, "use_scalar": false, "use_tucker": false, "use_wandb": false, "v2": false, "v_parameterization": false, "v_pred_like_loss": 0, "vae": "/root/kohya_ss/models/sdxl_vae_fp16fix.safetensors", "vae_batch_size": 0, "wandb_api_key": "", "wandb_run_name": "", "weighted_captions": false, "xformers": "xformers" }
FAQ
Comments (7)
What's a "coneja" anyways?
Ejaculation in condom. This trigger word set to prevent lora overlearning which fill your image with cum even there is no cum in your prompt.
Can you up date SD1.5 modle?
The dataset of PonyXLone and SD1.5 one are totally the same. The improvement is achieved by the change of base model. So there will not be an update for SD1.5.
It already exists, and it did when you made this comment.
Does it work with male characters?