model: illu0.1 <img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/7edf8d2c-eb21-4f2c-b8c8-3f2daa14908d/original=true/7edf8d2c-eb21-4f2c-b8c8-3f2daa14908d.jpeg" /> ob illu eps <img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/cc63a6df-778d-45bb-b4af-a53bb9c005c5/original=true/cc63a6df-778d-45bb-b4af-a53bb9c005c5.jpeg" /> ob vp <img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/5b4c559c-cb3c-4501-b976-153ac1bb2f92/original=true/5b4c559c-cb3c-4501-b976-153ac1bb2f92.jpeg" />vixson <img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/6f1ba9c2-1db8-4334-aa68-0aae2cfc8fbd/original=true/6f1ba9c2-1db8-4334-aa68-0aae2cfc8fbd.jpeg" /> dvine <img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/f401d58c-fd70-4754-a159-797add1d7ffa/original=true/f401d58c-fd70-4754-a159-797add1d7ffa.jpeg" /> i didnt ahve check the lora yet but this is the the result of other lora results what lora is used? lora: not the locon . new lyryko model: vixsion <img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/a60db5a5-41af-4679-bccd-cf247bdd3660/original=true/a60db5a5-41af-4679-bccd-cf247bdd3660.jpeg" /> model:vixsion <img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/6c5a9f91-bda6-4707-be91-aec0857e2d43/original=true/6c5a9f91-bda6-4707-be91-aec0857e2d43.jpeg" /> model:dvine <img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/f2f2bffc-ffec-4454-8269-24c541964f18/original=true/f2f2bffc-ffec-4454-8269-24c541964f18.jpeg" /> model:illu0.1 ofc the lora is trainned on this but it give me shit results note for me :never try to train on this lora type again<img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/b4b5fad1-1ea1-448e-af67-d9bd71d2f95a/original=true/b4b5fad1-1ea1-448e-af67-d9bd71d2f95a.jpeg" /><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/55d27b2a-0e7f-46b4-a83b-fc79e5822ee4/original=true/55d27b2a-0e7f-46b4-a83b-fc79e5822ee4.jpeg" /><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/f9576b89-8f92-4724-ad04-0b0beec19d81/original=true/f9576b89-8f92-4724-ad04-0b0beec19d81.jpeg" /><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/70186b08-a3e3-44b5-a40c-2f051cc28d34/original=true/70186b08-a3e3-44b5-a40c-2f051cc28d34.jpeg" />

marin style | trained on illu 0.1 |

<pre><code>

[network_arguments]</code></pre><pre><code>unet_lr = 0.0003</code></pre><pre><code>text_encoder_lr = 6e-5</code></pre><pre><code>network_dim = 64</code></pre><pre><code>network_alpha = 64</code></pre><pre><code>network_module = "networks.lora"</code></pre><pre><code>network_args = [ "conv_dim=64", "conv_alpha=64",]</code></pre><pre><code>network_train_unet_only = false</code></pre><pre><code>[optimizer_arguments]</code></pre><pre><code>learning_rate = 0.0003</code></pre><pre><code>lr_scheduler = "cosine_with_restarts"</code></pre><pre><code>lr_scheduler_num_cycles = 3</code></pre><pre><code>lr_warmup_steps = 350</code></pre><pre><code>optimizer_type = "AdamW8bit"</code></pre><pre><code>optimizer_args = [ "weight_decay=0.1", "betas=[0.9,0.99]",]</code></pre><pre><code>[training_arguments]</code></pre><pre><code>pretrained_model_name_or_path = "uYouUs/IllustriousV01"</code></pre><pre><code>vae = "stabilityai/sdxl-vae"</code></pre><pre><code>max_train_steps = 7000</code></pre><pre><code>train_batch_size = 4</code></pre><pre><code>seed = 42</code></pre><pre><code>max_token_length = 225</code></pre><pre><code>xformers = false</code></pre><pre><code>sdpa = true</code></pre><pre><code>min_snr_gamma = 8.0</code></pre><pre><code>lowram = true</code></pre><pre><code>no_half_vae = true</code></pre><pre><code>gradient_checkpointing = true</code></pre><pre><code>gradient_accumulation_steps = 1</code></pre><pre><code>max_data_loader_n_workers = 3</code></pre><pre><code>persistent_data_loader_workers = true</code></pre><pre><code>mixed_precision = "fp16"</code></pre><pre><code>full_bf16 = false</code></pre><pre><code>full_fp16 = false</code></pre><pre><code>cache_latents = true</code></pre><pre><code>cache_latents_to_disk = true</code></pre><pre><code>cache_text_encoder_outputs = false</code></pre><pre><code>min_timestep = 0</code></pre><pre><code>max_timestep = 1000</code></pre><pre><code>prior_loss_weight = 1.0</code></pre><pre><code>multires_noise_iterations = 6</code></pre><pre><code>multires_noise_discount = 0.3</code></pre><pre><code>v_parameterization = false</code></pre><pre><code>scale_v_pred_loss_like_noise_pred = false</code></pre><pre><code>zero_terminal_snr = false</code></pre><pre><code>[saving_arguments]</code></pre><pre><code>save_precision = "fp16"</code></pre><pre><code>save_model_as = "safetensors"</code></pre><pre><code>save_every_n_epochs = 1</code></pre><pre><code>save_last_n_epochs = 10</code></pre><pre><code>output_name = "64illu0lora"</code></pre><pre><code>output_dir = "/content/drive/MyDrive/Loras/example/output"</code></pre><pre><code>log_prefix = "64illu0lora"</code></pre><pre><code>logging_dir = "/content/Loras/_logs"</code></pre><pre><code>wandb_api_key = "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"</code></pre><pre><code>log_with = "wandb"</code></pre><pre><code>
</code></pre>

Files with SHA256: c8b837341228344fece5bd77014f1de43c0291015e834ed3714d11c77c2e0fb0