README.md

---
license: mit
base_model:
- StyleGAN
pipeline_tag: unconditional-image-generation
---

## 1. Overview

**StyleGAN-VectorFind-v6,v7** models for [Oh-LoRA v2 Project](https://github.com/WannaBeSuperteur/AI_Projects/tree/main/2025_05_02_OhLoRA_v2).

* These models are Fine-Tuned version of [GenForce StyleGAN celeba_partial-256x256](https://github.com/genforce/genforce/blob/master/MODEL_ZOO.md).
* Finally decided to use **StyleGAN-VectorFind-v7 (```stylegan_gen_vector_find_v7.pth```)**.
* Detailed info
  * [StyleGAN-VectorFind-v6 Detailed info (in Korean)](https://github.com/WannaBeSuperteur/AI_Projects/tree/main/2025_05_02_OhLoRA_v2/stylegan#3-2-stylegan-finetune-v1-%EA%B8%B0%EB%B0%98-%ED%95%B5%EC%8B%AC-%EC%86%8D%EC%84%B1%EA%B0%92-%EB%B3%80%ED%99%98-latent-z-vector-%ED%83%90%EC%83%89-stylegan-vectorfind-v6)
  * [StyleGAN-VectorFind-v7 Detailed info (in Korean)](https://github.com/WannaBeSuperteur/AI_Projects/tree/main/2025_05_02_OhLoRA_v2/stylegan#3-3-stylegan-finetune-v1-%EA%B8%B0%EB%B0%98-%ED%95%B5%EC%8B%AC-%EC%86%8D%EC%84%B1%EA%B0%92-%EB%B3%80%ED%99%98-intermediate-w-vector-%ED%83%90%EC%83%89-stylegan-vectorfind-v7)

## 2. Model files & save path for usage

All models have the same save path for usage : ```2025_05_02_OhLoRA_v2/stylegan/models/{model_file_name}```.

* Generator

|Version|Model|Finally decided to use|
|---|---|---|
|StyleGAN-VectorFind-v6|```stylegan_gen_vector_find_v6.pth```||
|StyleGAN-VectorFind-v7|```stylegan_gen_vector_find_v7.pth```|✅|

* Conv-NN for property score computation

|Version|Model type|Model|Finally decided to use|
|---|---|---|---|
|StyleGAN-FineTune-v2|CNN (for property score computation)|```stylegan_gen_fine_tuned_v2_cnn.pth```|✅|