RRIFT_notebooks

Paper: https://onlinelibrary.wiley.com/doi/full/10.1002/mrm.27913
Code: https://github.com/MPUmri/RRIFT
Note: MATLAB. Reproduces ALL figures

Setup instructions

Prerequisites

Make sure before running the notebook you download simResults.mat, simResultsTRes15-varKtRR.mat, simResultsTRes15-varVeRR.mat and simMap.mat from the OSF repository and place them inside ./RRIFT/data/. Also make sure you have Anaconda 3 installed on your system as well as MATLAB R2020b.

Installing MATLAB

You can download MATLAB from here.

Making a conda environment

• conda create -n rrift_notebook python=3.6 (you can set rrift_notebook to whatever you want the name of your environment to be)
• conda activate rrift_notebook

Installing the MATLAB kernel for Jupyter

• pip install matlab_kernel

To check if the kernel is properly installed use the command jupyter kernelspec list which should list both MATLAB and Python if installed correctly.

MATLAB-side configuration

The MATLAB executable needs to be exposed to Jypiter. You need to go to the directory where MATLAB is installed and navigate inside extern/engines/python. For example, the path should look something like this on Windows: M:\MATLAB\extern\engines\python or on Linux: /usr/local/MATLAB/R2020b/extern/engines/python.

Once you've navigated to that location you need to install the Python engine with the command:

• python setup.py install

Setting up SoS and MATLAB

• conda install sos sos-pbs -c conda-forge
• conda install sos-notebook jupyterlab-sos sos-papermill -c conda-forge
• conda install sos-python sos-matlab -c conda-forge

Installing Python packages

• pip install plotly=4.14.3 numpy=1.19.5 scipy=1.6.0 statsmodels=0.12.1 pandas=1.2.0

Start a Jupyter session

• jupyter notebook