训练样本的优化和提纯 ROI
为了克服LS-SVM解非稀疏性的缺点,同时抑制偏差较大的训练样本对海底趋势面构造的影响,提出一种基于局部样本中心距离的训练样本优化方法。
In order to solve the sparseness of LS-SVM results and restrain the influence of the sample-outliers, a new method of optimized samples by part samples center distance is presented.
我还要用到这里描述的样本用例,通过减少远程调用的数量优化实例的性能。
I will use the sample use case described here and optimize its performance by reducing the number of remote calls.
模糊数学方法的引用,使该模型不但能处理矛盾样本,而且有信息优化处理的功能。
Using fuzzy method we can not only treat with contradictory samples, but also make it in optimal case.
应用推荐