ClimSim
收藏arXiv2024-02-07 更新2024-08-06 收录
下载链接：
http://arxiv.org/abs/2306.08754v5
下载链接
链接失效反馈官方服务：
资源简介：
ClimSim是一个专为混合ML-物理学研究设计的大型多尺度数据集，包含57亿对多元输入和输出向量，用于隔离局部嵌套的高分辨率、高保真物理对主气候模拟器宏观物理状态的影响。该数据集具有全球覆盖，跨越多年，采样频率高，设计使得生成的模拟器与下游耦合到操作气候模拟器兼容。
ClimSim is a large-scale multi-scale dataset specifically designed for hybrid ML-physics research. It contains 5.7 billion pairs of multivariate input and output vectors, which are utilized to isolate the impact of locally nested, high-resolution and high-fidelity physical processes on the macroscopic physical states of a core climate simulator. This dataset features global coverage, spans multiple years, and has a high sampling frequency. It is engineered such that the resulting simulator is compatible with downstream coupling to operational climate simulators.
创建时间：
2023-06-15
