C ++库中的快速梯度下降实现? [英] Fast gradient-descent implementation in a C++ library?

问题描述

我正在寻找一种梯度下降优化方法，以最大程度地减少实例化变量的成本.我的程序在计算上非常昂贵，因此我正在寻找一种具有GD快速实现的流行库.推荐的库/参考是什么?

I'm looking to run a gradient descent optimization to minimize the cost of an instantiation of variables. My program is very computationally expensive, so I'm looking for a popular library with a fast implementation of GD. What is the recommended library/reference?

推荐答案

GSL 是优秀(免费)的库，已经实现了数学和科学兴趣的通用功能.

GSL is a great (and free) library that already implements common functions of mathematical and scientific interest.

您可以阅读整个在线参考手册.随便看看，这开始出现有趣，但是我认为我们需要更多地了解这个问题.

You can peruse through the entire reference manual online. Poking around, this starts to look interesting, but I think we'd need to know more about the problem.

这篇关于C ++库中的快速梯度下降实现?的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！