Install Python and MNE-Python


Users who work at a facility with a site-wide install of MNE (e.g. Martinos center) are encouraged to contact their technical staff about how to access and use MNE, as the instructions might differ.

1. Install a Python interpreter and dependencies

There are multiple options available for getting a suitable Python interpreter running on your system. However, for a fast and up to date scientific Python environment that resolves all dependencies, we recommend the Anaconda Python distribution.

Python has two major versions currently available, 2.7+ and 3.3+. Currently the 3D visualization dependencies Mayavi and PySurfer only run out of the box on Python 2.7, so we recommend using Python 2.7. You can get Anaconda 2.7 for Windows, OSX, and Linux download and installation instructions from the ContinuumIO site.

Once everything is set up, you should be able to check the version of conda and python that is installed:

$ conda --version
conda 4.2.14
$ which python
$ python --version
Python 2.7.12 :: Continuum Analytics, Inc.


If your installation doesn’t look something like this, something went wrong and you should try to fix it. Try looking through the Anaconda documentation or Googling for Anaconda install tips (StackExchange results are often helpful).

You can then do this to resolve the MNE dependencies:

$ conda install scipy matplotlib scikit-learn mayavi ipython-notebook
$ pip install PySurfer

Now that you have a working Python environment you can install MNE.

Users who would like a MATLAB-like interface should consider using Spyder, which can easily be installed $ conda install spyder.

2. Install the MNE module

There are a many options for installing MNE, but two of the most useful and common are:

  1. Use the stable release version of MNE. It can be installed as:

    $ pip install mne --upgrade

    We tend to release about once every six months, and this command can be used to update the install after each release.

  1. Use the development master version of MNE. If you want to be able to update your version between releases for bugfixes or new features, this will set you up for frequent updates:

    $ git clone git://
    $ cd mne-python
    $ python develop

    A feature of python develop is that any changes made to the files (e.g., by updating to latest master) will be reflected in mne as soon as you restart your Python interpreter. So to update to the latest version of the master development branch, you can do:

    $ git pull origin master

    and MNE will be updated to have the latest changes.

If you plan to contribute to MNE, please read how to Contribute to MNE.

3. Check your installation

To check that everything went fine, in ipython, type:

>>> import mne

If you get a new prompt with no error messages, you should be good to go!

A good place to start is on our Tutorials page or with our Examples Gallery. You can launch a web browser to the documentation with:

>>> mne.open_docs()  

Along the way, make frequent use of Python API Reference and User Manual to understand the capabilities of MNE.

For advanced topics like how to get NVIDIA CUDA support working for ~10x faster filtering and resampling, or if you’re having trouble, visit Advanced setup and troubleshooting.