A Python surrogate modeling framework with derivatives
A Python surrogate modeling framework with derivatives
复制标题
DOI：
10.1016/j.advengsoft.2019.03.005
复制
发表时间：
2019-09-01
影响因子：
4.8
通讯作者：
Martins, Joaquim R. R. A.
中科院分区：
文献类型：
--
作者：
Bouhlel, Mohamed Amine;Hwang, John T.;Martins, Joaquim R. R. A.
代理建模工具箱(SMT)是一个开源的Python包，其中包含代理建模方法、采样技术和基准测试函数的集合。此包提供了一个代理模型库，该库简单易用，并便于实现其他方法。SMT不同于现有的代理建模库，因为它强调导数，包括用于梯度增强建模的训练导数、预测导数和关于训练数据的导数。它还包括独特的代理模型：通过偏最小二乘减少的克立格法，它与输入的数量很好地缩放；以及能量最小化的样条内插，它与训练点的数量很好地缩放。通过一系列算例验证了SMT的有效性和有效性。SMT使用自定义工具进行记录，用于嵌入自动测试的代码和动态生成的曲线图，以最大限度地减少贡献者的工作，生成高质量的用户指南。SMT在公共版本控制存储库中维护。(1)
The surrogate modeling toolbox (SMT) is an open-source Python package that contains a collection of surrogate modeling methods, sampling techniques, and benchmarking functions. This package provides a library of surrogate models that is simple to use and facilitates the implementation of additional methods. SMT is different from existing surrogate modeling libraries because of its emphasis on derivatives, including training derivatives used for gradient-enhanced modeling, prediction derivatives, and derivatives with respect to training data. It also includes unique surrogate models: kriging by partial least-squares reduction, which scales well with the number of inputs; and energy-minimizing spline interpolation, which scales well with the number of training points. The efficiency and effectiveness of SMT are demonstrated through a series of examples. SMT is documented using custom tools for embedding automatically tested code and dynamically generated plots to produce high-quality user guides with minimal effort from contributors. SMT is maintained in a public version control repository.(1)