SUTD-PRCM
收藏资源简介:
SUTD-PRCM数据集由新加坡科技与设计大学创建,包含约260,000个复杂超表面的样本,这些样本通过电磁模拟生成。数据集中的每个样本包括一个16x16的二进制超表面图像及其相关的频率从2到10 GHz的电磁反射输出。该数据集被用于基准测试深度学习模型,特别适用于超表面的正向预测和逆向设计。通过将超表面图案分为不同类别,数据集支持不同复杂度的电磁响应分析,适用于评估和改进机器学习模型在超表面设计中的应用。
The SUTD-PRCM dataset was developed by the Singapore University of Technology and Design, consisting of approximately 260,000 complex metasurface samples generated via electromagnetic simulations. Each sample in the dataset includes a 16×16 binary metasurface image and its corresponding electromagnetic reflection responses over the frequency range of 2 to 10 GHz. This dataset serves as a benchmark for deep learning models, and is specifically tailored for forward prediction and inverse design of metasurfaces. By classifying metasurface patterns into distinct categories, the dataset enables electromagnetic response analysis across varying complexity levels, and is suitable for evaluating and enhancing the application of machine learning models in metasurface design.




