How to Use This Model
USE THEM SIMULTANEOUSLY. In this case, you need to download both ishtar_fgo.pt and
ishtar_fgo.safetensors, then use ishtar_fgo.pt as texture inversion embedding, and use
ishtar_fgo.safetensors as LoRA at the same time.
それらを同時に使用してください。この場合、ishtar_fgo.ptとishtar_fgo.safetensorsの両方をダウンロード
する必要があります。ishtar_fgo.ptをテクスチャ反転埋め込みとして使用し、同時にishtar_fgo.safetensorsをLoRAとして使用してください。
同时使用它们。在这种情况下，您需要下载ishtar_fgo.pt和ishtar_fgo.safetensors这两个文件，然后将ishtar_fgo.pt用作纹理反转嵌入，
同时使用ishtar_fgo.safetensors作为LoRA。
(Translated with ChatGPT)
The trigger word is ishtar_fgo, and the recommended tags are masterpiece, best quality, highres, solo, {ishtar_fgo:1.10}, long_hair, black_hair, two_side_up, red_eyes, jewelry, earrings, breasts, hoop_earrings, ribbon, bangs, smile, single_thighhigh, medium_breasts, hair_ribbon, armlet, parted_bangs, navel.
How This Model Is Trained
This model is trained with HCP-Diffusion. And the auto-training framework is maintained by DeepGHS Team.
Why Some Preview Images Not Look Like Ishtar Fgo
All the prompt texts used on the preview images (which can be viewed by clicking on the images) are automatically generated using clustering algorithms based on feature information extracted from the training dataset. The seed used during image generation is also randomly generated, and the images have not undergone any selection or modification. As a result, there is a possibility of the mentioned issues occurring.
In practice, based on our internal testing, most models that experience such issues perform better in actual usage than what is seen in the preview images. The only thing you may need to do is fine-tune the tags you use.
I Felt This Model May Be Overfitting or Underfitting, What Shall I Do
Our model has been published on huggingface repository - CyberHarem/ishtar_fgo, where models of all the steps are saved. Also, we published the training dataset on huggingface dataset - CyberHarem/ishtar_fgo, which may be help to you.
Why Not Just Using The Better-Selected Images
Our model's entire process, from data crawling, training, to generating preview images and publishing, is 100% automated without any human intervention. It's an interesting experiment conducted by our team, and for this purpose, we have developed a complete set of software infrastructure, including data filtering, automatic training, and automated publishing. Therefore, if possible, we would appreciate more feedback or suggestions as they are highly valuable to us.
Description
FAQ
Comments (2)
Please properly name your lora files, its hard to track if one has it or not if it has same name. Both v1.0 and v1.4 have the same filename.
Naming them like ishtar_fgo_narugo-v10.safetensors and ishtar_fgo_narugo-v14.safetensors would help much to reduce confusion with same named loras.
Thanks
Thank you for your feedback, but we don't think that falls within the scope of our concern.
We use the current naming convention precisely so that users can download and use the model without having to modify file names. Using a naming convention like 'character_series_narugo1992_v10' would actually create more trouble for the majority of users who don't have the same need for model archiving as you do.
Therefore, we suggest that if you do indeed need to manage a large number of different model versions, you should consider renaming the models yourself and organizing file paths in a structured way for retrieval.
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.