Dataset: ADM neural simulator for "Inverse deep learning methods and benchmarks for artificial electromagnetic material design"

The neural simulator for the ADM dataset of "Inverse deep learning methods and benchmarks for artificial electromagnetic material design"
For the usage, please refer to github page: https://github.com/BensonRen/AEM_DIM_Bench
Note that this is not a .csv format dataset but a neural simulator (model weights and definition of neural networks) since for inverse benchmarking, a simple dataset of paired geometry and spectra is not enough as one need to verify the designed geometry output from inverse algorithms (the forward mapping). The ADM dataset is one of the three dataset in the paper and the only one requiring a neural simulator due to its complexity.