Pre-print can be found on bioRxiv Documentation can be found here.
pip install git+https://github.com/funkelab/quacClone this repository, then install uv and run:
uv syncThis creates a .venv with the package and all dev dependencies.
Run tests with uv run pytest, and linting with uv run prek run -a.
To install the git hooks, run uv run prek install.
On linux, the default torch wheels from PyPI include CUDA.
To use smaller CPU-only wheels instead, run uv sync --group cpu.
Alternatively, you can use pixi
(e.g. pixi shell -e dev).
Logo made with the help of DALL-E 2. The code for training the StarGAN is adapted from the Official Pytorch Implementation of StarGANv2.