PySEF: A python library for similarity-based dimensionality reduction

作者：

Highlights：

• Open-source implementation of the similarity embedding framework.

• Supports both CPU-based and GPU-based optimization.

• Optimized for handling large-scale datasets.

• Provides a number of pre-implemented similarity targets.

• Allows for easily defining novel dimensionality reduction techniques.

摘要

•Open-source implementation of the similarity embedding framework.•Supports both CPU-based and GPU-based optimization.•Optimized for handling large-scale datasets.•Provides a number of pre-implemented similarity targets.•Allows for easily defining novel dimensionality reduction techniques.

论文关键词：Dimensionality reduction,Similarity embedding framework

论文评审过程：Received 23 October 2017, Revised 5 April 2018, Accepted 8 April 2018, Available online 21 April 2018, Version of Record 12 May 2018.

论文官网地址：https://doi.org/10.1016/j.knosys.2018.04.019