Installation guide

Requirements

Python ≥3.10, <3.13
uv package manager (recommended) or pip

If you do not have uv installed:


curl -LsSf https://astral.sh/uv/install.sh | sh

Install

Create and activate a virtual environment first:

macOS / Linux


uv venv
source .venv/bin/activate

Then install the base package (includes core data handling, classical ML, gradient-boosted trees (XGBoost, LightGBM, CatBoost), hyperparameter optimisation (Optuna), and visualisation):


uv pip install -e .

Optional extras

Extra	Command	Notes
`dl`	`uv pip install -e ".[dl]"`	Torch package that enambles deep learning architectures to be run
`xai`	`uv pip install -e ".[xai]"`	Explainable AI packages including SHAP, LIME, PyALE
`app`	`uv pip install -e ".[app]"`	Running web app can be run only when FastAPI and uvicorn packages are installed
`fl`	`uv pip install -e ".[fl]"`	Federated learning dependencies including flower
`tab-fm`	`uv pip install -e ".[tab-fm]"`	Tabular foundation models including TabPFN, TabICL
`llm`	`uv pip install -e ".[llm]"`	Ollama for running classification with large language models configured as classifiers
`automl`	`uv pip install -e ".[automl]"`	FLAML for automated search over learners
`privacy`	`uv pip install -e ".[privacy]"`	Opacus (differentially-private training)
`mgm`	`uv pip install -e ".[mgm]"`	Microbial general model (MGM) as encoder https://github.com/HUST-NingKang-Lab/MGM
`dev`	`uv pip install -e ".[dev]"`	pytest, black, isort, flake8, mypy, jupyter
`all`	`uv pip install -e ".[all]"`	Everything above

Verify installation


import mllabiome as mll
print(mll.__version__)

GPU support (optional)

For CUDA support, replace the torch install with the appropriate wheel from pytorch.org .