Install

Setting up BenchNIRS

1. Download and install Python 3.9 or greater, for example with Miniconda.

2. To install the package with pip (cf. PyPI), open a terminal (eg. Anaconda Prompt) and type:

   pip install benchnirs
   

3. Download the datasets (see below).

Note

Alternatively to install from source in development mode, download and unzip the repository (or clone it with Git), and run devinstall.py.

Downloading the datasets

• Herff et al. 2014 (n-back task)

• Shin et al. 2018 (n-back and word generation tasks)

• Shin et al. 2016 (mental arithmetic task, select NIRS_01-29 for the fNIRS data)

• Bak et al. 2019 (motor execution task)

Enabling GPU support

To enable GPU acceleration on BenchNIRS with PyTorch (if your machine has an NVIDIA GPU), open a terminal (eg. Anaconda Prompt) and type:

pip install --upgrade torch --index-url https://download.pytorch.org/whl/cu118

For more information please see the PyTorch documentation.

Keeping BenchNIRS up to date

To update BenchNIRS to the latest version with pip, open a terminal (eg. Anaconda Prompt) and type:

pip install --upgrade benchnirs