This repository primarily provides a Gradio GUI for Kohya's Stable Diffusion trainers. However, support for Linux OS is also offered through community contributions. macOS support is not optimal at the moment but might work if the conditions are favorable.
The GUI allows you to set the training parameters and generate and run the required CLI commands to train the model.
This Colab notebook was not created or maintained by me; however, it appears to function effectively. The source can be found at: https://github.com/camenduru/kohya_ss-colab.
I would like to express my gratitude to camendutu for their valuable contribution. If you encounter any issues with the Colab notebook, please report them on their repository.
Run one of the following setup script by executing the following command:
For systems with only python 3.10.11 installed:
.\setup.bat
For systems with only more than one python release installed:
.\setup-3.10.bat
During the accelerate config step, use the default values as proposed during the configuration unless you know your hardware demands otherwise. The amount of VRAM on your GPU does not impact the values used.
Optional: CUDNN 8.9.6.50
The following steps are optional but will improve the learning speed for owners of NVIDIA 30X0/40X0 GPUs. These steps enable larger training batch sizes and faster training speeds.
Run .\setup.bat and select 2. (Optional) Install cudnn files (if you want to use the latest supported cudnn version).
Linux and macOS
Linux Pre-requirements
To install the necessary dependencies on a Linux system, ensure that you fulfill the following requirements:
Ensure that venv support is pre-installed. You can install it on Ubuntu 22.04 using the command:
apt install python3.10-venv
Install the CUDA 11.8 Toolkit by following the instructions provided in this link.
Make sure you have Python version 3.10.9 or higher (but lower than 3.11.0) installed on your system.
Setup Linux
To set up the project on Linux or macOS, perform the following steps:
Open a terminal and navigate to the desired installation directory.
Clone the repository by running the following command:
If you encounter permission issues, make the setup.sh script executable by running the following command:
chmod +x ./setup.sh
Run the setup script by executing the following command:
./setup.sh
Note: If you need additional options or information about the runpod environment, you can use setup.sh -h or setup.sh --help to display the help message.
Install Location
The default installation location on Linux is the directory where the script is located. If a previous installation is detected in that location, the setup will proceed there. Otherwise, the installation will fall back to /opt/kohya_ss. If /opt is not writable, the fallback location will be $HOME/kohya_ss. Finally, if none of the previous options are viable, the installation will be performed in the current directory.
For macOS and other non-Linux systems, the installation process will attempt to detect the previous installation directory based on where the script is run. If a previous installation is not found, the default location will be $HOME/kohya_ss. You can override this behavior by specifying a custom installation directory using the -d or --dir option when running the setup script.
If you choose to use the interactive mode, the default values for the accelerate configuration screen will be "This machine," "None," and "No" for the remaining questions. These default answers are the same as the Windows installation.
Runpod
Manual installation
To install the necessary components for Runpod and run kohya_ss, follow these steps:
Select the Runpod pytorch 2.0.1 template. This is important. Other templates may not work.
SSH into the Runpod.
Clone the repository by running the following command:
To upgrade your installation to a new version, follow the instructions below.
Windows Upgrade
If a new release becomes available, you can upgrade your repository by running the following commands from the root directory of the project:
Pull the latest changes from the repository:
git pull
Run the setup script:
.\setup.bat
Linux and macOS Upgrade
To upgrade your installation on Linux or macOS, follow these steps:
Open a terminal and navigate to the root directory of the project.
Pull the latest changes from the repository:
git pull
Refresh and update everything:
./setup.sh
Starting GUI Service
To launch the GUI service, you can use the provided scripts or run the kohya_gui.py script directly. Use the command line arguments listed below to configure the underlying service.
text
1--listen: Specify the IP address to listen on for connections to Gradio.
2--username: Set a username for authentication.
3--password: Set a password for authentication.
4--server_port: Define the port to run the server listener on.
5--inbrowser: Open the Gradio UI in a web browser.
6--share: Share the Gradio UI.
7--language: Set custom language
Launching the GUI on Windows
On Windows, you can use either the gui.ps1 or gui.bat script located in the root directory. Choose the script that suits your preference and run it in a terminal, providing the desired command line arguments. Here's an example:
The repository now provides a default configuration file named config.toml. This file is a template that you can customize to suit your needs.
To use the default configuration file, follow these steps:
Copy the config example.toml file from the root directory of the repository to config.toml.
Open the config.toml file in a text editor.
Modify the paths and settings as per your requirements.
This approach allows you to easily adjust the configuration to suit your specific needs to open the desired default folders for each type of folder/file input supported in the GUI.
You can specify the path to your config.toml (or any other name you like) when running the GUI. For instance: ./gui.bat --config c:\my_config.toml
LoRA
To train a LoRA, you can currently use the train_network.py code. You can create a LoRA network by using the all-in-one GUI.
Once you have created the LoRA network, you can generate images using auto1111 by installing this extension.
Sample image generation during training
A prompt file might look like this, for example:
txt
1# prompt 1
2masterpiece, best quality, (1girl), in white shirts, upper body, looking at viewer, simple background --n low quality, worst quality, bad anatomy, bad composition, poor, low effort --w 768 --h 768 --d 1 --l 7.5 --s 28
34# prompt 2
5masterpiece, best quality, 1boy, in business suit, standing at street, looking back --n (low quality, worst quality), bad anatomy, bad composition, poor, low effort --w 576 --h 832 --d 2 --l 5.5 --s 40
Lines beginning with # are comments. You can specify options for the generated image with options like --n after the prompt. The following options can be used:
--n: Negative prompt up to the next option.
--w: Specifies the width of the generated image.
--h: Specifies the height of the generated image.
--d: Specifies the seed of the generated image.
--l: Specifies the CFG scale of the generated image.
--s: Specifies the number of steps in the generation.
The prompt weighting such as ( ) and [ ] is working.
Troubleshooting
If you encounter any issues, refer to the troubleshooting steps below.
Page File Limit
If you encounter an X error related to the page file, you may need to increase the page file size limit in Windows.
No module called tkinter
If you encounter an error indicating that the module tkinter is not found, try reinstalling Python 3.10 on your system.
LORA Training on TESLA V100 - GPU Utilization Issue
Issue Summary
When training LORA on a TESLA V100, users reported low GPU utilization. Additionally, there was difficulty in specifying GPUs other than the default for training.
Potential Solutions
GPU Selection: Users can specify GPU IDs in the setup configuration to select the desired GPUs for training.
Improving GPU Load: Utilizing adamW8bit optimizer and increasing the batch size can help achieve 70-80% GPU utilization without exceeding GPU memory limits.
SDXL training
The documentation in this section will be moved to a separate document later.
Masked loss
The masked loss is supported in each training script. To enable the masked loss, specify the --masked_loss option.
The feature is not fully tested, so there may be bugs. If you find any issues, please open an Issue.
ControlNet dataset is used to specify the mask. The mask images should be the RGB images. The pixel value 255 in R channel is treated as the mask (the loss is calculated only for the pixels with the mask), and 0 is treated as the non-mask. The pixel values 0-255 are converted to 0-1 (i.e., the pixel value 128 is treated as the half weight of the loss). See details for the dataset specification in the LLLite documentation.
