DeepXDE Benchmark Dataset
收藏arXiv2025-09-30 收录
下载链接：
https://github.com/stfc-sciml/ZeroCoordinateShift
下载链接
链接失效反馈官方服务：
资源简介：
该数据集旨在基准测试ZCS算法与其他自动微分策略在解决高阶线性偏微分方程时的性能表现。它被生成用于比较不同自动微分策略在训练过程中的GPU内存消耗和墙钟时间。数据集包含了不同数量的函数（M）、配置点（N）以及最大微分阶数（P）的变化规模，以支持进行物理信息操作学习基准分析的任务。
This dataset is designed to benchmark the performance of the ZCS algorithm and other automatic differentiation strategies when solving high-order linear partial differential equations. It is generated to compare the GPU memory consumption and wall-clock time of different automatic differentiation strategies during training. The dataset includes varying scales of the number of functions (M), collocation points (N), and maximum differential order (P), to support benchmark analysis tasks for physics-informed operator learning.
提供机构：
DeepXDE
