ICA with MNE-Python

Installation

Installing conda

If you don't have python, I recommand to install it using conda. Conda is an open-source package management system and environment management system that runs on Windows, macOS, and Linux. Conda quickly installs, runs, and updates packages and their dependencies. Conda easily creates, saves, loads, and switches between environments on your local computer.

For a quick and light installation, I recommand to use miniconda a free minimal installer for conda.

Select the version corresponding to your operating system and download and install it.

Create a new environment

Once completed, you should have access to a new shell named Anaconda Prompt on your computer.

Launch it and create a new python environment using the conda create command.

conda create -n eeg39 python=3.9

Here we specify the name of the new environment eeg39 and the python version 3.9.

Anaconda will ask you if you are sure you can to create this new environment

Proceed ([y]/n)?

Press y (yes) then Enter to accept

Installing dependencies

Activate the new environment using:

conda activate eeg39

Notice that the current environment is displayed at the beginning of your shell:

(eeg39) C:\Users\user_name>

Install the required python packages:

• notebook to use jupyter notebook interface:

  pip install notebook

• mne the main python EEG librairy.

  pip install mne

• picard an implementation of the picard ICA solver.

  pip install python-picard

• mne-qt-browser a 2D backend for plotting MNE data.

  pip install PyQt5 mne matplotlib mne-qt-browser

• pybv to export file to the brainvision format

  pip install pybv

• pandas library to work with dataframes

  pip install pandas

• pycartool library to open cartool files.

  pip install pycartool

To install everything at once:

pip install notebook mne python-picard matplotlib PyQt5 mne-qt-browser pybv pandas pycartool

or

pip install -r requirements.txt

Use the code

Each time you want to use the environment, you need to activate it using:

conda activate eeg39

Then start the jupyter server:

jupyter notebook