更新 function.py

function.py

@@ -27,11 +27,14 @@ name = "章程鸿"
 
 # 这里写自己的代码实现
 import numpy as np
+from sklearn.metrics import accuracy_score
 
+# 定义sigmoid函数
 def sigmoid(z):
     return 1 / (1 + np.exp(-z))
 
-def train_logreg(X, y, learning_rate, iterations):
+# 训练逻辑回归模型
+def train_logreg(X, y, learning_rate, iterations=1000):
     m, n = X.shape
     weights = np.zeros(n)
     bias = 0
@@ -42,13 +45,15 @@ def train_logreg(X, y, learning_rate, iterations):
         db = (1/m) * np.sum(h - y)
         weights -= learning_rate * dw
         bias -= learning_rate * db
-    return weights
+    return weights, bias
 
-def predict_logreg(inX, weights):
-    z = np.dot(inX, weights)
+# 使用逻辑回归模型进行预测
+def predict_logreg(inX, weights, bias):
+    z = np.dot(inX, weights) + bias
     h = sigmoid(z)
     return 1 if h >= 0.5 else 0
 
+# 加载数据
 def load_data(filepath):
     data = []
     with open(filepath, 'r') as file:
@@ -64,24 +69,36 @@ def load_data(filepath):
     y = data[:, -1]
     return X, y
 
+# 寻找最佳学习率
+def find_best_learning_rate(X, y, learning_rates, iterations=1000):
+    best_accuracy = 0
+    best_learning_rate = None
+    best_weights = None
+    best_bias = None
+    for rate in learning_rates:
+        weights, bias = train_logreg(X, y, rate, iterations)
+        y_pred = np.array([predict_logreg(x, weights, bias) for x in X])
+        accuracy = accuracy_score(y, y_pred)
+        if accuracy > best_accuracy:
+            best_accuracy = accuracy
+            best_learning_rate = rate
+            best_weights = weights
+            best_bias = bias
+    return best_learning_rate, best_accuracy, best_weights, best_bias
+
 if __name__ == "__main__":
     X_train, y_train = load_data("./data/horse-colic.data")
     X_test, y_test = load_data("./data/horse-colic.test")
-    best_learning_rate = None
-    best_accuracy = 0
-    learning_rates = [0.001, 0.002, 0.003, 0.004, 0.005, 0.006, 0.007, 0.008, 0.009, 0.01]
-    iterations = 1000
-    for lr in learning_rates:
-        weights = train_logreg(X_train, y_train, learning_rate=lr, iterations=iterations)
-        correct_count = 0
-        for i in range(len(X_test)):
-            y_pred = predict_logreg(X_test[i], weights)
-            if y_pred == y_test[i]:
-                correct_count += 1
-        accuracy = correct_count / len(X_test)
-        if accuracy > best_accuracy:
-            best_accuracy = accuracy
-            best_learning_rate = lr
-    best_weights = train_logreg(X_train, y_train, learning_rate=best_learning_rate, iterations=iterations)
-    y_pred = predict_logreg(X_test[11], best_weights)
-    print(f"预测结果: {y_pred}, 实际结果: {y_test[11]}")
+    
+    # 定义要尝试的学习率列表
+    learning_rates = np.linspace(0.001, 0.1, 50)
+    
+    # 寻找最佳学习率
+    best_learning_rate, best_accuracy, best_weights, best_bias = find_best_learning_rate(X_train, y_train, learning_rates)
+    
+    print(f"最佳学习率: {best_learning_rate}")
+    print(f"最佳准确率: {best_accuracy}")
+    
+    # 使用最佳权重和偏置进行预测
+    y_pred = predict_logreg(X_test[11], best_weights, best_bias)
+    print(f"预测结果: {y_pred}, 实际结果: {y_test[11]}")