# Fine-Tuning A Model with Kohya

## How to Fine-Tuning A Model with Kohya <a href="#id-8a1e" id="id-8a1e"></a>

### Build LoRA Models for Stable Diffusion <a href="#id-5d9f" id="id-5d9f"></a>

**Hugging Face LoRA Library:** [**https://huggingface.co/lora-library**](https://huggingface.co/lora-library)\
**CivitAI:** [**https://civitai.com**](https://civitai.com/)&#x20;

**GitHub Kohya**: [**https://github.com/bmaltais/kohya\_ss**](https://github.com/bmaltais/kohya_ss)

## <https://www.youtube.com/watch?v=xholR62Q2tY>

## Train a LoRA Model

&#x20;[02:06](https://www.youtube.com/watch?v=xholR62Q2tY\&t=126s) Build your database

&#x20;[02:48](https://www.youtube.com/watch?v=xholR62Q2tY\&t=168s) Training images

&#x20;[04:22](https://www.youtube.com/watch?v=xholR62Q2tY\&t=262s) Captioning (BLIP) [09:57](https://www.youtube.com/watch?v=xholR62Q2tY\&t=597s) Regularization images

&#x20;[13:10](https://www.youtube.com/watch?v=xholR62Q2tY\&t=790s) Choose your base model for training

&#x20;[16:09](https://www.youtube.com/watch?v=xholR62Q2tY\&t=969s) Create folders to train the model

&#x20;[21:48](https://www.youtube.com/watch?v=xholR62Q2tY\&t=1308s) Theory: What is a Neural Network?

&#x20;[23:40](https://www.youtube.com/watch?v=xholR62Q2tY\&t=1420s) Theory: Batch

&#x20;[24:50](https://www.youtube.com/watch?v=xholR62Q2tY\&t=1490s) Theory: Epoch

&#x20;[25:20](https://www.youtube.com/watch?v=xholR62Q2tY\&t=1520s) Theory: Iterations, or Steps, or Repeat

&#x20;[26:00](https://www.youtube.com/watch?v=xholR62Q2tY\&t=1560s) Theory: Loss Function

&#x20;[26:49](https://www.youtube.com/watch?v=xholR62Q2tY\&t=1609s) Theory: Learning Rate

&#x20;[27:43](https://www.youtube.com/watch?v=xholR62Q2tY\&t=1663s) Set the key training parameters

&#x20;[32:12](https://www.youtube.com/watch?v=xholR62Q2tY\&t=1932s) How to use a LoRA modell