# Installation

## Stable release

To install icomo, run this command in your
terminal:

```bash
$ pip install icomo
```

It might be the case that the system doesn't find the graphviz software installed via
pip. This is needed to display the flows between compartment In this case follow the
installation instructions from the [graphviz website](https://graphviz.org/download/). 

To run all the examples, you might also want to install the optional dependencies:

```bash
$ pip install icomo[opt]
```

You might want to create a virtual environment first, e.g. using
[mamba](https://mamba.readthedocs.io/en/latest/mamba-installation.html):

```bash
$ mamba create -n icomo_env icomo
$ mamba activate icomo_env
```

or directly using venv:

```bash
$ python -m venv /path/to/new/icomo_env
$ source /path/to/new/icomo_env/bin/activate
$ pip install icomo
```

You might also want to install [jupyterlab](https://jupyter.org) to run the examples:

```bash
$ pip install jupyterlab
```

## From sources

The sources for icomo can be downloaded from
the [Github repo](https://github.com/Priesemann-Group/icomo).

You can clone the public repository:

```bash
$ git clone https://github.com/Priesemann-Group/icomo
```

Enter the directory and install with pip:

```bash
$ cd icomo
$ pip install -e .
```
This enables you to edit the code and have the changes directly available in your python
environment.
To include the optional and development dependencies, use:

```bash
$ cd icomo
$ pip install -e .[dev]
```