We recommend using python3.6 and pycharm.
We recommend resetting a new environment for cst-python-co-simulatioin, eg:python3.6(cst_env)
Create a new system environment variable named PYTHONPATH and set its value to “AMD64\python_cst_libraries” in the CST installation directory.
The project is mainly used to randomly generate and simulate meta-atoms on demand, and establish datasets that can be used to train and test the neural networks. Also, the metasurface is modeled automatically in cst according to the geometric parameters.
If you use this code for your research, please cite:
Spectral-transfer-learning-based Metasurface Design Methodology Assisted by Complex-valued Deep Neural Network