xs

Author: sheng.xu

.idea/workspace.xml

@@ -197,4 +197,4 @@
 4.15244831176753	1.44597290703838	1
 9.55986996363196	1.13832040773527	1
 1.63276516895206	0.446783742774178	1
-9.38532498107474	0.913169554364942	1
+9.38532498107474	0.913169554364942	1

Chapter_1 Logistic Regression/data.txt

@@ -26,9 +26,6 @@ def load_data(file_name):
         feature_data.append(feature_tmp)
         label_data.append(lable_tmp)
     f.close()  # 关闭文件
-    print feature_data
-    print "======"
-    print label_data
     return np.mat(feature_data), np.mat(label_data)
 
 if __name__ == "__main__":
@@ -38,4 +35,3 @@ def load_data(file_name):
     # 2、训练LR模型
     print feature
     print label
-    print "---------- 2.training ------------"

Chapter_1 Logistic Regression/load_test.py

@@ -78,6 +78,7 @@ def save_result(file_name, result):
     print "---------- 1.load model ------------"
     w = load_weight("weights")
     n = np.shape(w)[1]
+    print np.shape(w)
     # 2、导入测试数据
     print "---------- 2.load data ------------"
     testData = load_data("test_data", n)
@@ -86,5 +87,5 @@ def save_result(file_name, result):
     h = predict(testData, w)#进行预测
     # 4、保存最终的预测结果
     print "---------- 4.save prediction ------------"
-    print np.mat(h)
-    save_result("result.model", np.mat(h))
+    #print np.mat(h)
+    save_result("result", np.mat(h))

Chapter_1 Logistic Regression/lr_test.py

@@ -44,7 +44,8 @@ def lr_train_bgd(feature, label, maxCycle, alpha):
     output: w(mat):权重
     '''
     n = np.shape(feature)[1]  # 特征个数
-    w = np.mat(np.ones((n, 1)))  # 初始化权重
+    # (200,3) np.shape(feature)[0] 表示特征矩阵的行数，np.shape(feature)[1] 表示特征矩阵的个数
+    w = np.mat(np.ones((n, 1)))  # 初始化权重  初始化三个1的一个矩阵
     i = 0
     while i <= maxCycle:  # 在最大迭代次数的范围内
         i += 1  # 当前的迭代次数
@@ -89,9 +90,12 @@ def save_model(file_name, w):
     # 1、导入训练数据
     print "---------- 1.load data ------------"
     feature, label = load_data("data.txt")
+    print feature
+    print label
     # 2、训练LR模型
     print "---------- 2.training ------------"
     w = lr_train_bgd(feature, label, 1000, 0.01)
+    print w
     # 3、保存最终的模型
     print "---------- 3.save model ------------"
     save_model("weights", w)

Chapter_1 Logistic Regression/lr_train.py

@@ -197,4 +197,4 @@
 2.74262556141998	-0.254027266166219	1
 1.24804345641085	-0.241408809531456	1
 0.808431917613837	-0.242017996831424	1
-6.74094601972670	-0.373633464603415	1
+6.74094601972670	-0.373633464603415	1