@@ -1126,6 +1126,39 @@ def selectThreshold(yval,pval):
11261126 若:
11271127 ![ enter description here] [ 55 ] ,
11281128 表示x1,x2** 负相关**
1129+ - 实现代码:
1130+ ```
1131+ # 多元高斯分布函数
1132+ def multivariateGaussian(X,mu,Sigma2):
1133+ k = len(mu)
1134+ if (Sigma2.shape[0]>1):
1135+ Sigma2 = np.diag(Sigma2)
1136+ '''多元高斯分布函数'''
1137+ X = X-mu
1138+ argu = (2*np.pi)**(-k/2)*np.linalg.det(Sigma2)**(-0.5)
1139+ p = argu*np.exp(-0.5*np.sum(np.dot(X,np.linalg.inv(Sigma2))*X,axis=1)) # axis表示每行
1140+ return p
1141+ ```
1142+ ### 6、单元和多元高斯分布特点
1143+ - 单元高斯分布
1144+ - 人为可以捕捉到` feature ` 之间的关系时可以使用
1145+ - 计算量小
1146+ - 多元高斯分布
1147+ - 自动捕捉到相关的feature
1148+ - 计算量大,因为:![ $$ \Sigma \in {R^{n \times {\rm{n}}}} $$ ] ( http://latex.codecogs.com/png.latex?%5Cfn_cm%20%24%24%5CSigma%20%5Cin%20%7BR%5E%7Bn%20%5Ctimes%20%7B%5Crm%7Bn%7D%7D%7D%7D%24%24 )
1149+ - ` m>n ` 或` Σ ` 可逆时可以使用。(若不可逆,可能有冗余的x,因为线性相关,不可逆,或者就是m<n)
1150+
1151+ ### 7、程序运行结果
1152+ - 显示数据
1153+ ![ enter description here] [ 56 ]
1154+ - 等高线
1155+ ![ enter description here] [ 57 ]
1156+ - 异常点
1157+ ![ enter description here] [ 58 ]
1158+
1159+
1160+
1161+ ----------------------------------
11291162
11301163
11311164 [ 1 ] : ./images/LinearRegression_01.png " LinearRegression_01.png "
@@ -1182,4 +1215,7 @@ def selectThreshold(yval,pval):
11821215 [ 52 ] : ./images/AnomalyDetection_03.png " AnomalyDetection_03.png "
11831216 [ 53 ] : ./images/AnomalyDetection_05.png " AnomalyDetection_05.png "
11841217 [ 54 ] : ./images/AnomalyDetection_07.png " AnomalyDetection_07.png "
1185- [ 55 ] : ./images/AnomalyDetection_06.png " AnomalyDetection_06.png "
1218+ [ 55 ] : ./images/AnomalyDetection_06.png " AnomalyDetection_06.png "
1219+ [ 56 ] : ./images/AnomalyDetection_08.png " AnomalyDetection_08.png "
1220+ [ 57 ] : ./images/AnomalyDetection_09.png " AnomalyDetection_09.png "
1221+ [ 58 ] : ./images/AnomalyDetection_10.png " AnomalyDetection_10.png "
0 commit comments