Yet Another Dataset Tagger
The recommended way of running the repo is using uv.
Once uv is installed, you can run uv sync --extra=cpu in order to grab the necessary Python dependencies. If you have an Nvidia GPU you can use uv sync --extra=cuda118 or uv sync --extra=cuda124 instead. For AMD GPUs, you can use uv sync --extra=rocm.
Afterwards, simply run uv run main.py in order to start the UI.
You can also use Python's built-in virtual environments to run the project. Run python3 -m venv .venv to create your environment, then activate it using . .venv/bin/activate. To install the dependencies, you need to run pip install -r requirements-cpu.txt (or the different requirements file, depending on your machine).
Afterwards, run python3 main.py to start the UI.
