📝 PCA主成分分析，数据降维

Author: lawlite19

PCA/PCA.py

@@ -102,6 +102,7 @@ def projectData(X_norm,U,K):
 def drawline(plt,p1,p2,line_type):
     plt.plot(np.array([p1[0],p2[0]]),np.array([p1[1],p2[1]]),line_type)  
 
+  
 
 # 恢复数据
 def recoverData(Z,U,K):

readme.md

@@ -949,14 +949,14 @@ def runKMeans(X,initial_centroids,max_iters,plot_process):
  - 又因为`Ureduce`为正定矩阵，【正定矩阵满足：![$$A{A^T} = {A^T}A = E$$](http://latex.codecogs.com/gif.latex?%5Cfn_cm%20%24%24A%7BA%5ET%7D%20%3D%20%7BA%5ET%7DA%20%3D%20E%24%24)，所以：![$${A^{ - 1}} = {A^T}$$](http://latex.codecogs.com/gif.latex?%5Cfn_cm%20%24%24%7BA%5E%7B%20-%201%7D%7D%20%3D%20%7BA%5ET%7D%24%24)】，所以这里：
  - ![$${X_{approx}} = {(U_{reduce}^{ - 1})^{ - 1}}Z = {U_{reduce}}Z$$](http://latex.codecogs.com/gif.latex?%5Cfn_cm%20%24%24%7BX_%7Bapprox%7D%7D%20%3D%20%7B%28U_%7Breduce%7D%5E%7B%20-%201%7D%29%5E%7B%20-%201%7D%7DZ%20%3D%20%7BU_%7Breduce%7D%7DZ%24%24)
  - 实现代码：
- ```
+```
     # 恢复数据 
     def recoverData(Z,U,K):
         X_rec = np.zeros((Z.shape[0],U.shape[0]))
         U_recude = U[:,0:K]
         X_rec = np.dot(Z,np.transpose(U_recude))  # 还原数据（近似）
         return X_rec
- ```
+```
 
 ### 6、主成分个数的选择（即要降的维度）
 - 如何选择