Polyglot MATLAB Notebooks

A repository containing examples and instructions on using Script of Script (SoS) Polyglot notebooks with MATLAB and other languages.

Polyglot MATLAB Notebooks using Script of Script (SoS)

A repository containing examples and instructions on using Script of Script (SoS) Polyglot notebooks with MATLAB and other languages.

Here are the tools I'm using

• OS: Windows 11
• git: 2.37.2.windows.2
• MATLAB: R2022b (9.13.0.2049777 )
• Terminal launching Miniconda3
  • Conda: 23.3.1
  • Python: 3.10.10
  • pip: 23.1.2

Installation Steps

1. Install and activate MATLAB

2. Install miniconda3 or anaconda3 (I use miniconda) and open anaconda terminal

• Note: Run the terminal in administrator or elevated permission (depending on your OS)

3. Create an environment for MATLAB in miniconda and activate

• conda create -n "sos" python=3.10

• conda activate sos

4. (In MATLAB console) Point MATLAB to your conda environment

• pyversion('C:\Users\<user>\miniconda3\envs\matlab\python.exe');
  pyenv('ExecutionMode', 'OutOfProcess');

5. Install SoS, Python and MATLAB subkernels and matplotlib (for Python) via

• run in conda terminal:

  conda install sos sos-pbs sos-notebook jupyterlab-sos sos-papermill sos-python sos-r matplotlib -c conda-forge

  pip install sos-matlab

• Source: https://vatlab.github.io/sos-docs/running.html#Conda-installation

6. Install the MATLAB engine to Python (note that official documentation is deprecated on MathWorks and SoS websites now)

• In conda terminal run:

python -m pip install matlabengine==9.13.7

• To test that this worked, first activate Python in the conda terminal via:

• python

• Then execute the following code:

  import matlab.engine
  eng = matlab.engine.start_matlab()
  eng.isprime(37)

  • This should return True. If there were errors, troubleshoot at https://github.com/mathworks/matlab-engine-for-python
• Exit Python with exit() but keep terminal open.

7. (still in conda terminal) Install matlab_kernel

• pip install matlab_kernel

8. Confirm that the installation works by running the example/test_installation notebook and running each cell.

• Open a Jupyter Notebook in conda terminal via:

jupyter notebook

Notes:

1. You will need to change the ratelimit for Jupyter to visualize larger EEGLAB objects. Follow the steps here:
   • https://stackoverflow.com/questions/43288550/iopub-data-rate-exceeded-in-jupyter-notebook-when-viewing-image
   • jupyter notebook --generate-config
     • edit this file
     • search for c.NotebookApp.iopub_data_rate_limit
     • uncomment this and set the value to: =1.0e10