UI/UX Note:
All CLI output in Dataset Forge uses the Catppuccin Mocha color scheme for a consistent, visually appealing experience. All prompts, menus, and progress bars are styled using the centralized color utilities.
Welcome to Dataset Forge!
This guide will help you install and launch Dataset Forge for the first time.
- Python: 3.12+ (see requirements.txt)
- OS: Windows (primary)
- CUDA/cuDNN: For GPU acceleration (see Special Installation)
- RAM: 8GB+ (16GB+ recommended)
- Storage: SSD recommended
-
Clone the repository:
git clone https://github.com/Courage-1984/Dataset-Forge.git cd Dataset-Forge -
Set up the environment:
Option A: Automated Installation (Recommended)
# Windows (easiest) install.bat # or tools\install.bat # Or manually py -3.12 -m venv venv312 venv312\Scripts\activate python tools\install.py
Option B: Manual Installation
py -3.12 -m venv venv312 venv312\Scripts\activate python -m pip install --upgrade pip setuptools wheel # Install the correct CUDA-enabled torch/torchvision/torchaudio first! pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu121 pip install -r requirements.txt pip install -e .
Option C: Using setup.py directly
py -3.12 -m venv venv312 venv312\Scripts\activate python setup.py install
Note: For other CUDA versions, see PyTorch Get Started.
./run.bat
OR
venv312\Scripts\activate
py main.py
OR
dataset-forge-
On Windows,
python-magicrequires extra DLLs. -
You must install VapourSynth before using getnative.
-
You must compile/build resdet before using it.
-
AND MORE;;;
-
See Special Installation Instructions for further details.
- For common issues, see the Troubleshooting Guide.
- For advanced configuration, see Advanced Features.