Installation

You can install CaliBrain using either pip or Conda, depending on your preferences.

We provide the following installation options:

Choose the method that best fits your workflow.

Pip Installation (Recommended)

The recommended way to install CaliBrain is using pip with the pyproject.toml file.

First, clone the repository:

git clone https://github.com/braindatalab/CaliBrain.git
cd CaliBrain

Then install the package:

pip install .

If you are actively developing the code and want automatic updates when you edit files:

pip install -e .

This will install all dependencies defined inside pyproject.toml.

Alternatively, you can install CaliBrain using a traditional requirements.txt:

pip install -r requirements.txt

Note:

This method is simpler but does not capture full metadata (e.g., Python version compatibility).
Make sure your environment uses a supported Python version (>=3.8).

If you prefer to manage dependencies with Conda, you can create an isolated Conda environment using the provided environment.yml file.

First, clone the repository:

git clone https://github.com/braindatalab/CaliBrain.git
cd CaliBrain

Then create the environment:

conda env create -f environment.yml

Activate the environment:

conda activate calibrain

Finally, install CaliBrain into the activated environment:

pip install .

This ensures that all Conda and pip dependencies are properly installed.

Recommended: Use pip with pyproject.toml for clean dependency management (pip install .).
If you prefer Conda: Use environment.yml to create a Conda environment first.
If you just want quick pip install: Use requirements.txt.

All methods lead to the same installed package — just choose the method that matches your ecosystem (pip-only or Conda).

Python >= 3.8
Tested on Python 3.8, 3.9, 3.10
Operating systems: Linux, macOS, Windows (WSL recommended for full compatibility)

If you plan to contribute to CaliBrain or run experiments:

pip install -e .[dev]

(Development dependencies will be added soon.)