From bc1dc96315e9e32e9eca98abf4ef7cd872959603 Mon Sep 17 00:00:00 2001
From: jiangzhonglian <jiang-s@163.com>
Date: Thu, 6 Apr 2017 23:15:28 +0800
Subject: [PATCH] =?UTF-8?q?=E6=9B=B4=E6=96=B0=2014.SVD=E6=B3=A8=E9=87=8A?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 src/python/14.SVD/svdRec.py | 17 ++++++++---------
 1 file changed, 8 insertions(+), 9 deletions(-)

diff --git a/src/python/14.SVD/svdRec.py b/src/python/14.SVD/svdRec.py
index a85fe180..c051a7c4 100644
--- a/src/python/14.SVD/svdRec.py
+++ b/src/python/14.SVD/svdRec.py
@@ -37,6 +37,7 @@ def loadExData2():
 
 
 def loadExData():
+    """
     # 推荐引擎示例矩阵
     return[[4, 4, 0, 2, 2],
            [4, 0, 0, 3, 3],
@@ -54,7 +55,7 @@ def loadExData():
            [1, 1, 0, 2, 2],
            [0, 0, 0, 3, 3],
            [0, 0, 0, 1, 1]]
-    """
+  
 
 
 # 相似度计算，假定inA和inB 都是列向量
@@ -211,18 +212,18 @@ def imgCompress(numSV=3, thresh=0.8):
 
 
 if __name__ == "__main__":
-    """
+
     # 对矩阵进行SVD分解(用python实现SVD)
     Data = loadExData()
+    print Data
     U, Sigma, VT = linalg.svd(Data)
     # 打印Sigma的结果，因为前3个数值比其他的值大了很多，为9.72140007e+00，5.29397912e+00，6.84226362e-01
     # 后两个值比较小，每台机器输出结果可能有不同可以将这两个值去掉
-    # print Sigma
+    print Sigma
 
     # 重构一个3x3的矩阵Sig3
-    # Sig3 = mat([[Sigma[0], 0, 0], [0, Sigma[1], 0], [0, 0, Sigma[2]]])
-    # print U[:, :3] * Sig3 * VT[:3, :]
-    """
+    Sig3 = mat([[Sigma[0], 0, 0], [0, Sigma[1], 0], [0, 0, Sigma[2]]])
+    print U[:, :3] * Sig3 * VT[:3, :]
 
     """
     # 计算欧氏距离
@@ -241,6 +242,7 @@ if __name__ == "__main__":
 
     """
 
+    """
     # 计算相似度的方法
     myMat = mat(loadExData2())
     # print myMat
@@ -249,7 +251,6 @@ if __name__ == "__main__":
     # 计算相似度的第二种方式
     print recommend(myMat, 1, estMethod=svdEst, simMeas=pearsSim)
 
-    """
     # 默认推荐（菜馆菜肴推荐示例）
     print recommend(myMat, 2)
     """
@@ -269,5 +270,3 @@ if __name__ == "__main__":
 
     # 压缩图片
     # imgCompress(2)
-
-