《machine learning》Andrew Ng chapter 13 support vector machine

1. optimization objection (优化目标)

1.1 SVM hypothesis:

SVM hypothesis
SVM
- cost1 与 cost0 均为代价函数;
- 左图用于正样本,右图适用于负样本;
- SVM / large margin classifier:用较大间距将样本区分开;
large margin classifier

  • 大间距分类器;
  • 若 number of features >> number of training examples,则使用 logistic regression, or SVM without a kernel(linear kernel);
    若 number of features < number of training examples,则使用 SVM with Gaussian kernel;
    若 number of features < number of training examples,则 create/add more features, then use logistic regression or SVM without a kernel;

refenence: 《machine learning》Andrew Ng

猜你喜欢

转载自blog.csdn.net/Maximize1/article/details/81514086