add code anomaly detection异常检测代码

lawlite19 · lawlite19 · commit 3bc2172d05b0 · 2016-12-28T14:37:05.000+08:00
diff --git a/AnomalyDetection/AnomalyDetection.py b/AnomalyDetection/AnomalyDetection.py
@@ -13,13 +13,25 @@ def anomalyDetection_example():
     plt = display_2d_data(X, 'bx')
     plt.title("origin data")
     plt.show()
+    '''多元高斯分布函数，并可视化拟合的边界'''
+    mu,sigma2 = estimateGaussian(X)         # 参数估计（求均值和方差）
+    #print mu,sigma2
+    p = multivariateGaussian(X,mu,sigma2)   # 多元高斯分布函数
+    #print p
+    visualizeFit(X,mu,sigma2)  # 显示图像
     
-    mu,sigma2 = estimateGaussian(X)
-    print mu,sigma2
-    p = multivariateGaussian(X,mu,sigma2)
-    print p
+    '''选择异常点（在交叉验证CV上训练得到最好的epsilon）'''
+    Xval = data['Xval']
+    yval = data['yval']
+    pval = multivariateGaussian(Xval, mu, sigma2) # 计算CV上的概率密度值
+    epsilon,F1 = selectThreshold(yval,pval)       # 选择最优的epsilon临界值
+    print u'在CV上得到的最好的epsilon是：%e'%epsilon
+    print u'对应的F1Score值为：%f'%F1
+    outliers = np.where(p<epsilon)  # 找到小于临界值的异常点，并作图
+    plt.plot(X[outliers,0],X[outliers,1],'o',markeredgecolor='r',markerfacecolor='w',markersize=10.)
+    plt = display_2d_data(X, 'bx')
+    plt.show()
     
-    visualizeFit(X,mu,sigma2)
     
     
     
@@ -44,16 +56,49 @@ def multivariateGaussian(X,mu,Sigma2):
     k = len(mu)
     if (Sigma2.shape[0]>1):
         Sigma2 = np.diag(Sigma2)
-        
+    '''多元高斯分布函数'''    
     X = X-mu
     argu = (2*np.pi)**(-k/2)*np.linalg.det(Sigma2)**(-0.5)
     p = argu*np.exp(-0.5*np.sum(np.dot(X,np.linalg.inv(Sigma2))*X,axis=1))  # axis表示每行
     return p
     
 # 可视化边界
 def visualizeFit(X,mu,sigma2):
-    X1,X2 = np.meshgrid(0,0.5,35)
-    Z = multivariateGaussian(np.vstack((X1,X2)), mu, Sigma2)
+    x = np.arange(0, 36, 0.5) # 0-36，步长0.5
+    y = np.arange(0, 36, 0.5)
+    X1,X2 = np.meshgrid(x,y)  # 要画等高线，所以meshgird
+    Z = multivariateGaussian(np.hstack((X1.reshape(-1,1),X2.reshape(-1,1))), mu, sigma2)  # 计算对应的高斯分布函数
+    Z = Z.reshape(X1.shape)  # 调整形状
+    plt.plot(X[:,0],X[:,1],'bx')
+    
+    if np.sum(np.isinf(Z).astype(float)) == 0:   # 如果计算的为无穷，就不用画了
+        # plt.contourf(X1,X2,Z,10.**np.arange(-20, 0, 3),linewidth=.5)
+        CS = plt.contour(X1,X2,Z,10.**np.arange(-20, 0, 3),color='black',linewidth=.5)   # 画等高线，Z的值在10.**np.arange(-20, 0, 3)
+        #plt.clabel(CS)
+            
+    plt.show()
+
+# 选择最优的epsilon，即：使F1Score最大    
+def selectThreshold(yval,pval):
+    '''初始化所需变量'''
+    bestEpsilon = 0.
+    bestF1 = 0.
+    F1 = 0.
+    step = (np.max(pval)-np.min(pval))/1000
+    '''计算'''
+    for epsilon in np.arange(np.min(pval),np.max(pval),step):
+        cvPrecision = pval<epsilon
+        tp = np.sum((cvPrecision == 1) & (yval == 1)).astype(float)  # sum求和是int型的，需要转为float
+        fp = np.sum((cvPrecision == 1) & (yval == 0)).astype(float)
+        fn = np.sum((cvPrecision == 1) & (yval == 0)).astype(float)
+        precision = tp/(tp+fp)  # 精准度
+        recision = tp/(tp+fn)   # 召回率
+        F1 = (2*precision*recision)/(precision+recision)  # F1Score计算公式
+        if F1 > bestF1:  # 修改最优的F1 Score
+            bestF1 = F1
+            bestEpsilon = epsilon
+    return bestEpsilon,bestF1
+        
 
 
 if __name__ == '__main__':
