简体繁体 English

Apache Commons数学优化

[英]Apache Commons Math optimization

原文 2010-04-27 00:08:43 9 2 java/ math/ optimization/ apache-commons-math

Does anyone have any experience with the Apache Commmons Math optimization package? 有没有人对Apache Commmons Math优化包有任何经验？ More specifically, the Nelder-Mead method implementation? 更具体地说，Nelder-Mead方法的实现？ Is it pretty high quality? 它质量很高吗？

2 个解决方案

Nelder-Mead is one of the most popular multivariate optimization algorithm. Nelder-Mead是最受欢迎的多变量优化算法之一。 The Apache implmentation is pretty good. Apache的实施非常好。

However, if you have more information, eg, gradient, you should consider using a more "informed" algorithm such as BFGS. 但是，如果您有更多信息，例如渐变，则应考虑使用更“明智”的算法，例如BFGS。 It also works with analytical gradient (using finite differencing). 它也适用于分析梯度（使用有限差分）。 R uses BFGS by default, I think... R默认使用BFGS，我认为......

SuanShu has implemented 10+ Java optimization algorithms that you can choose for your purpose. SuanShu已经实现了10多种Java优化算法，您可以根据自己的需要进行选择。 Hope this helps. 希望这可以帮助。

Note that Apache Commons Math also contains Michael Powell's direct search BOBYQA algorithm. 请注意，Apache Commons Math还包含Michael Powell的直接搜索BOBYQA算法。 This algorithm generally converges in substantially fewer iterations than the classical Nelder-Mead method, plus it also supports bounds on the variables. 该算法通常收敛在比经典内尔德-米德法迭代基本上更少的， 再加上它也支持对变量界限。