Matlab

Especially in engineering disciplines, Matlab by Mathworks is a popular scripting languange for numerical simulations. While Hermes is a Python tool, almost completely implemented in Python and C++, it is possible to control modules and routines from Matlab’s Python interface, esentially integrating Hermes’ full feature set into existing Matlab workflows. Hermes requires Python 3.9, the respective Matlab interface requires a specific Matlab Version.

Assuming Hermes has been properly installed on the host system, before the execution of any python matlab routines, the virtual environment has to be selected:

pyenv('Version', '<path>', 'ExecutionMode', 'OutOfProcess');

Note that path has to point to the python.exe of your virtual environment. So, for instance assuming Conda has been used to create Hermes’ virtual Python environment, <path> should look something like C:\Users\<username>\.conda\envs\<envname>\python, <username> and <envname> being replaced by the respective directory names.

After environment activation, any Hermes module can be loaded, objects initialized and functions called. As a minimal example, initializing a new simulation and adding a single virtual device can be achieved via

simulation_module = py.importlib.import_module('hermespy.simulation');
simulation = simulation_module.Simulation();
device = simulation.new_device();

Several complete examples are hosted within the official GitHub repository. They should provide a starting point for anyone interested in integrating Hermes functionalities into their Matlab workflow. Thanks to Roberto Bomfin for providing the source files. Please note that Hermes does not officially support complete Matlab integration of all features. Instead, this documentation is only served as a courtesy to users and developers.