daebakgazua's picture
DOC: add base model & pipeline tag
4374463 verified
---
license: mit
base_model:
- StyleGAN
pipeline_tag: unconditional-image-generation
---
## 1. Overview
**StyleGAN-VectorFind-v9 & additional** models for [Oh-LoRA v3.1 Project](https://github.com/WannaBeSuperteur/AI_Projects/tree/main/2025_06_07_OhLoRA_v3_1).
* StyleGAN-FineTune-v9 models are Fine-Tuned version of [GenForce StyleGAN celeba_partial-256x256](https://github.com/genforce/genforce/blob/master/MODEL_ZOO.md).
* [Detailed info (in Korean)](https://github.com/WannaBeSuperteur/AI_Projects/blob/main/2025_06_07_OhLoRA_v3_1/stylegan/README.md)
## 2. Model files & save path for usage
* **Property Score CNN (Merged)** is a merged CNN model [(Model Structure PDF)](https://github.com/WannaBeSuperteur/AI_Projects/blob/main/2025_05_26_OhLoRA_v3/property_score_cnn/model_structure_pdf/merged_property_cnn.pdf), of CNN models below:
* [Property Score CNN for computing ```eyes``` (눈 뜬 정도), ```mouth``` (입 벌린 정도), ```pose``` (고개 돌림), ```hair_color``` (머리 색), ```hair_length``` (머리 길이), and ```background_mean``` (배경색 평균 밝기) properties](https://github.com/WannaBeSuperteur/AI_Projects/tree/main/2025_04_08_OhLoRA/stylegan_and_segmentation#3-3-cnn-model-%EB%82%98%EB%A8%B8%EC%A7%80-%ED%95%B5%EC%8B%AC-%EC%86%8D%EC%84%B1-%EA%B0%92-7%EA%B0%9C)
* Property Score CNN for computing ```hairstyle``` (직모 vs. 곱슬) property value [(Ground truth value 계산 알고리즘)](https://github.com/WannaBeSuperteur/AI_Projects/blob/main/2025_05_26_OhLoRA_v3/stylegan/README.md#2-2-%EC%A7%81%EB%AA%A8-vs-%EA%B3%B1%EC%8A%AC%EB%A8%B8%EB%A6%AC-%EC%86%8D%EC%84%B1-%EA%B0%92-%EA%B3%84%EC%82%B0-%EC%95%8C%EA%B3%A0%EB%A6%AC%EC%A6%98)
* [What is 'Property Score' (핵심 속성 값)?](https://github.com/WannaBeSuperteur/AI_Projects/blob/main/2025_06_07_OhLoRA_v3_1/stylegan/README.md#2-%ED%95%B5%EC%8B%AC-%EC%86%8D%EC%84%B1-%EA%B0%92)
|Model|Model File|Save Path for use<br>(Directory)|
|---|---|---|
|Property Score CNN (Merged)|```ohlora_v3_merged_property_cnn.pth```|```2025_06_07_OhLoRA_v3_1/property_score_cnn/models```|
|[StyleGAN-FineTune-v9](https://github.com/WannaBeSuperteur/AI_Projects/blob/main/2025_06_07_OhLoRA_v3_1/stylegan/README.md#3-2-fine-tuned-stylegan-stylegan-finetune-v9) Generator|```stylegan_gen_fine_tuned_v9.pth```|```2025_06_07_OhLoRA_v3_1/stylegan/models```|
|[StyleGAN-FineTune-v9](https://github.com/WannaBeSuperteur/AI_Projects/blob/main/2025_06_07_OhLoRA_v3_1/stylegan/README.md#3-2-fine-tuned-stylegan-stylegan-finetune-v9) Discriminator|```stylegan_dis_fine_tuned_v9.pth```|```2025_06_07_OhLoRA_v3_1/stylegan/models```|
|StyleGAN-VectorFind-v9 [(option: ```mixed```)](https://github.com/WannaBeSuperteur/AI_Projects/blob/main/2025_06_07_OhLoRA_v3_1/stylegan/stylegan_vectorfind_v9/image_generation_report.md#1-1-%EC%B5%9C%EC%A2%85-%EC%9D%B4%EB%AF%B8%EC%A7%80-%EC%83%9D%EC%84%B1-%ED%85%8C%EC%8A%A4%ED%8A%B8-option-1-mixed) Generator|```stylegan_gen_vector_find_v9 (mixed).pth```|```2025_06_07_OhLoRA_v3_1/stylegan/models```|
|[Simple Neural Network to get gradient for ```eyes``` property change vector](https://github.com/WannaBeSuperteur/AI_Projects/tree/main/2025_06_07_OhLoRA_v3_1/stylegan#3-4-stylegan-vectorfind-v9-gradient-%EB%B0%A9%EB%B2%95) for StyleGAN-VectorFind-v9 [(option: ```mixed```)](https://github.com/WannaBeSuperteur/AI_Projects/blob/main/2025_06_07_OhLoRA_v3_1/stylegan/stylegan_vectorfind_v9/image_generation_report.md#1-1-%EC%B5%9C%EC%A2%85-%EC%9D%B4%EB%AF%B8%EC%A7%80-%EC%83%9D%EC%84%B1-%ED%85%8C%EC%8A%A4%ED%8A%B8-option-1-mixed) Generator|```stylegan_gen_vector_find_v9_nn_eyes (mixed).pth```|```2025_06_07_OhLoRA_v3_1/stylegan/models```|
|StyleGAN-VectorFind-v9 [(option: ```svm_ms2```)](https://github.com/WannaBeSuperteur/AI_Projects/blob/main/2025_06_07_OhLoRA_v3_1/stylegan/stylegan_vectorfind_v9/image_generation_report.md#1-2-%EC%B5%9C%EC%A2%85-%EC%9D%B4%EB%AF%B8%EC%A7%80-%EC%83%9D%EC%84%B1-%ED%85%8C%EC%8A%A4%ED%8A%B8-option-2-svm_ms2) Generator|```stylegan_gen_vector_find_v9 (svm_ms2).pth```|```2025_06_07_OhLoRA_v3_1/stylegan/models```|