#核treelets(kt)
KernelTreelets的Python项目详细描述
核树是由Hedi Xia和Hector D.Ceniceros提出的一种分层聚类算法。 它将数据的一种特殊的多尺度分解treelets与再生核hilbert空间上的投影相结合。 该方法称为核treelets(kt),有效地将treelets中使用的相关系数矩阵替换为对称的。 由核函数有效构造半正定矩阵。 与大多数需要数据集是数字的聚类方法不同, kt可以应用于更一般的数据,直接在特征空间中生成基于数据的多分辨率序列。 arxiv(https://arxiv.org/abs/1812.04808)中提供了关于内核树的更详细的解释。 这篇论文被数据科学和自适应分析的进展所接受。 最新的测试版本位于https://github.com/hedixia/KernelTreelets_v5。
此回购协议仅用于测试上述文章中的示例。 由于安装文件尚未完成,实现可能尚未完全优化,它仍在构建中。 当前,可以通过将主文件夹中扩展名为.py的所有文件复制到执行文件的路径来使用它。
有关内核方法的更多信息,请参见(https://en.wikipedia.org/wiki/Kernel_method)。
有关ann b.lee、boaz nadler和larry wasserman的treelets的更多信息,请参见arxiv(https://arxiv.org/abs/0707.0481)。
主页:https://github.com/hedixia/KernelTreelets_v5 作者:夏海霞 作者电子邮件:hedixia@ucsb.edu 许可证:未知 说明:内核treelets(kt)
Kernel Treelets is a hierarchical clustering algorithm proposed by Hedi Xia and Hector D. Ceniceros. It combines treelets, a particular multiscale decomposition of data, with a projection on a reproducing kernel Hilbert space. The proposed approach, called kernel treelets (KT), effectively substitutes the correlation coefficient matrix used in treelets with a symmetric, positive semi-definite matrix efficiently constructed from a kernel function. Unlike most clustering methods, which require data sets to be numeric, KT can be applied to more general data and yield a multi-resolution sequence of basis on the data directly in feature space. A more detailed explanation about Kernel Treelets can be found in arXiv (https://arxiv.org/abs/1812.04808). The paper is accepted by Advances in Data Science and Adaptive Analysis. The most recent test version is at https://github.com/hedixia/KernelTreelets_v5.
This repo is only for testing the examples in the paper above. The implementation may not be fully optimized and it is still under construction as the setup files are not yet finished. Currently it can be used by copying all files in the main folder with .py extensions to the path of the executed file.
For more infomation about Kernel Method, see (https://en.wikipedia.org/wiki/Kernel_method).
For more infomation about Treelets by Ann B. Lee, Boaz Nadler and Larry Wasserman, see arXiv (https://arxiv.org/abs/0707.0481).
平台:未知 分类器:编程语言::python::3 分类器:许可证::OSI批准::MIT许可证 分类器:操作系统::与操作系统无关