Install via pip or conda

Braindecode requires two libraries already installed before being installed, PyTorch and moabb if you want to use MOABB datasets utilities.

Preparing the environment for braindecode installation

We suggest installing the Pytorch into its own conda or pip environment. Installing the PyTorch library depends on the architecture and operational system you are running, as well as driver versions and optimizations for deep learning.

Given this complexity, we recommend that you go directly to the pytorch page and follow the instructions in http://pytorch.org/. You don’t need to install torchvision.

To install moabb, run in your terminal:

$ pip install moabb

Installing the braindecode with pip

After preparing the environment for braindecode, you can install via pip:

$ pip install braindecode

alternatively, you can install the latest version of braindecode via pip:

pip install -U https://api.github.com/repos/braindecode/braindecode/zipball/master

Installing Braindecode with conda

You can also create a conda environment for installing the library. This might help depending on how you are managing other libraries.

curl -O https://raw.githubusercontent.com/braindecode/braindecode/master/environment.yml
conda env create -f environment.yml
conda activate braindecode

This will create a new conda environment called braindecode (you can adjust this by passing a different name via --name).