************
Installation
************

Requirements
============

To use `mGrowthCtrl`, you will need `Python`_ and a set of required packages. For optimal compatibility and performance, we recommend using the following minimum versions:

* Python_      ``>= 3.10``
* numpy_       ``>= 1.26,<2.0``
* scipy_       ``>= 1.10``
* pandas_      ``>= 2.0``
* matplotlib_  ``>= 3.7``
* PyTorch_     ``>= 2.5`` (optional, for neural controllers)
* torchdiffeq_ ``>= 0.2`` (optional, for neural ODE solving)
* pymcmcstat_  ``>= 1.9`` (optional, for MCMC fitting)

The required packages will be installed automatically during the installation process according to ``pyproject.toml``.


Install `mGrowthCtrl`
=====================

To install the package locally, clone the repository and use pip in a virtual
environment:

.. code-block:: bash

   git clone https://gitlab.com/ComputationalScience/mgrowthctrl.git
   cd mgrowthctrl

   python -mvenv .venv
   source .venv/bin/activate

   pip install -e .

For optional features (neural controllers and MCMC), install with extras:

.. code-block:: bash

   pip install -e ".[torch,mcmc]"

Alternatively, you can use the uv_ package manager and follow the instructions
in their documentation.

.. _Python:      https://www.python.org/downloads/
.. _numpy:       https://numpy.org/install/
.. _scipy:       https://scipy.org/install/
.. _pandas:      https://pandas.pydata.org/
.. _matplotlib:  https://matplotlib.org/stable/users/getting_started/
.. _PyTorch:     https://pytorch.org/get-started/locally/
.. _torchdiffeq: https://github.com/rtqichen/torchdiffeq
.. _pymcmcstat:  https://pymcmcstat.readthedocs.io/
.. _uv:          https://docs.astral.sh/uv/