CSOptim

Gradient-based optimization of core-shell particles with distinct material classes. For detailed information please check the article

Requirements

Setup a suitable conda environment named cs_optim with:

conda env create -f environment.yaml
conda activate cs_optim

Run Optimization

You can run an Optimization with the main.py file. The code includes the generation of target spectra of bare core and core-single-shell particles with PyMieScatt. To target single or mutliple frequencies add pass them as -enhancement and add linewidths with -width.

Models and Training

The trained models can be found here. To start your own training, use main_train.py with your prefered parameters. You may generate your own training data using PyMieScat. The original training data used for the models is available here.