iivs-cardio

A monorepo for deep-learning and bio-imaging research on cardiomyocytes

📦 Development setup

Requires Python 3.14+.

git clone https://github.com/iivs-lab/cardiomyocytes.git
cd cardiomyocytes
uv sync --group dev

🖥️ Compute environment

This project targets PyTorch on NVIDIA GPUs:

• Python 3.14+
• PyTorch 2.12.1 / torchvision 0.27.1
• Compute backend: CUDA 13.0 — wheels are pulled from the dedicated PyTorch index (https://download.pytorch.org/whl/cu130) configured in pyproject.toml; uv sync installs them automatically.

An NVIDIA GPU with a CUDA 13.0-capable driver is expected. The first uv sync downloads the CUDA build of torch (~1.8 GB).

Switching the compute backend

The backend is selected by the [[tool.uv.index]] named pytorch in pyproject.toml:

• CUDA: set the URL to .../whl/cu126, .../whl/cu128, or .../whl/cu130.
• CPU-only: set the URL to .../whl/cpu.

Then re-resolve with uv lock && uv sync. Alternatively re-run copier update --UNSAFE --vcs-ref pytorch and re-answer compute_backend / cuda_version.

📋 TODO

See TODO.md for tracked open items.

📜 Changelog

See CHANGELOG.md for the version history.

⚖️ License

This project is distributed under the terms of the MIT license.