LibSVM 是SVM分類和回歸問題的有效解決方案。這里的以php接口封裝的svm擴展是為了在php腳本中更方便的應用 支持向量機(*SVM*)號稱最優秀的分類算法之一 [https://pecl.php.net/package/svm](https://pecl.php.net/package/svm) * [例子](https://www.php.net/manual/en/svm.examples.php) * [SVM](https://www.php.net/manual/en/class.svm.php)— SVM類 * [SVM :: \_\_ construct](https://www.php.net/manual/en/svm.construct.php)—構造一個新的SVM對象 * [SVM :: crossvalidate](https://www.php.net/manual/en/svm.crossvalidate.php)—在訓練數據的子集上測試訓練參數 * [SVM :: getOptions](https://www.php.net/manual/en/svm.getoptions.php)—返回當前的訓練參數 * [SVM :: setOptions](https://www.php.net/manual/en/svm.setoptions.php)—設置訓練參數 * [SVM :: train](https://www.php.net/manual/en/svm.train.php)—基于訓練數據創建SVMModel * [SVMModel](https://www.php.net/manual/en/class.svmmodel.php)— SVMModel類 * [SVMModel :: checkProbabilityModel](https://www.php.net/manual/en/svmmodel.checkprobabilitymodel.php)—如果模型具有概率信息，則返回true * [SVMModel :: \_\_ construct](https://www.php.net/manual/en/svmmodel.construct.php)—構造一個新的SVMModel * [SVMModel :: getLabels](https://www.php.net/manual/en/svmmodel.getlabels.php)—獲取在其上訓練模型的標簽 * [SVMModel :: getNrClass](https://www.php.net/manual/en/svmmodel.getnrclass.php)—返回使用模型訓練的類的數量 * [SVMModel :: getSvmType](https://www.php.net/manual/en/svmmodel.getsvmtype.php)—獲取模型訓練所使用的SVM類型 * [SVMModel :: getSvrProbability](https://www.php.net/manual/en/svmmodel.getsvrprobability.php)—獲取回歸類型的sigma值 * [SVMModel :: load](https://www.php.net/manual/en/svmmodel.load.php)—加載保存的SVM模型 * [SVMModel :: predict\_probability](https://www.php.net/manual/en/svmmodel.predict-probability.php)—先前看不見的數據的返回類概率 * [SVMModel :: predict](https://www.php.net/manual/en/svmmodel.predict.php)—預測以前看不見的數據的值 * [SVMModel :: save](https://www.php.net/manual/en/svmmodel.save.php)—將模型保存到文件