Please read through the entire description (might need to be expanded) and the version change notes as they cover a lot of information about basic use cases and limitations. The description is always focused on the latest version. Thank you!This is an SD 1.5 LoRA for the character Serena from Pokemon. It works for models based on SD 1.x only, it will not work for models based on SD 2.x or SDXL!Example images were picked from pure 512x512 txt2img results and then re-created at 1024x1024 using txt2img with moderate Hi-res Fix settings (upscaler Lanczos, denoising strength 0.4). This improves faces and other details that are almost impossible to get correctly and consistently at 512x512 (limitation of the technology) while still giving a realistic impression of what it looks like. Pure 512x512 results will have more distorted faces and less detail.Please see the version change notes for the training and example image generation models as well as the used weights as they might change between versions. Remember that you might need to adjust weights to best suit your use case!Remember to add the tag serena \(pokemon\) (with backslashes intact) to your positive prompt.The training set contained multiple variations of Serena's signature look (from the game, not the anime, sorry) so you might need to put combinations of the following tags in your positive or negative prompts to get the desired results:<ul><li>hat</li><li>eyewear on headwear</li><li>sleeveless shirt</li><li>high-waist skirt</li><li>thighhighs</li><li>shoes</li><li>mega ring</li><li>shoulder bag</li></ul>As the training set contained a few images with floating hair, you might need to add floating hair to the negative prompts if not desired.Known Limitations / Problems:<ul><li>This model is only trained on the game design for Serena, not the anime. I'm currently not interested to do the anime design, sorry.</li><li>The shoes and bag were not present in a lot of images and not a focus for the training so they sadly will not be consistent at all.</li><li>The face, eyes and mega ring will similarly be very inconsistent because of a lot of different interpretations in the training set, especially for wider shots. The eye color should be specified explicitly in the positive prompts, e.g. grey eyes.</li><li>Depending on the image composition (angle, distance etc.), it might be more or less difficult to remove the hat. Both hat and eyewear on headwear in the negative prompts and no headwear in the positive prompts should help but you still may need some luck.</li><li>Even with floating hair in the negative prompts, it tends to go a bit wild. To be fair, that's true for many images I've seen as well.</li><li>Shots from behind are very inconsistent due to extremely limited training data. Might for instance generate wrong hair, not apply poses correctly etc. Not much I can do without a bit more training data.</li></ul>

[SD 1.5] Pokemon - Serena (Game)

General changes:<ul><li>Changed example weight from ~0.5 to ~0.7 (result of training changes)</li><li>Removed three images from the training set that did not fit the style I am going for</li></ul>Changes to the training process (with the goal of hopefully making this and my other models more flexible and compatible while reducing overfitting):<ul><li>Switched from a flat 200 repeats x 1 epoch to 20 repeats x 10 epochs (max) and selecting the best results</li><li>Lowered network training alpha from 128 to 64 to hopefully increase the quality of the result a bit</li><li>Resized LoRA from rank 128 to rank 64 rank after training (still training with rank 128) to reduce size and "smooth out" some of the training results while not changing much (going down to rank 32 seems to have too much influence on the quality of results, so I kept it at 64)</li></ul>Used weight: 0.7Training model: <a target="_blank" rel="ugc" href="https://civitai.com/models/66/anything-v3">Anything V3</a>Example image generation model: <a target="_blank" rel="ugc" href="https://civitai.com/models/4451/abyssorangemix2-hardcore">AbyssOrangeMix2 - Hardcore</a>

Files with SHA256: f4d79fbd2434355244618b526c778d64fa6ef7b4d37803fdcf96e197614575cc