更新 function.py

function.py

@@ -30,7 +30,7 @@ def sigmoid(z):
     return 1.0 / (1 + np.exp(-z))
 
 
-def train_logreg(X, y, learning_rate=0.01, iterations=1000, batch_size=32, regularization=None, lambda_reg=0.1):
+def train_logreg(X, y, learning_rate=0.05, iterations=1500, batch_size=64, regularization='L2', lambda_reg=0.05):
     """
     对率回归训练函数。
 
@@ -124,53 +124,12 @@ def load_data(filepath):
 
     return data, target_variable
 
-def accuracy(X, y, weights):
-    predictions = [predict_logreg(sample, weights) for sample in X]
-    return np.mean(predictions == y)  # 计算准确率
-
-def grid_search(X_train, y_train, X_test, y_test, param_grid):
-    best_accuracy = 0
-    best_params = None
-
-    for params in product(*param_grid.values()):
-        learning_rate, iterations, batch_size, regularization, lambda_reg = params
-
-        # 训练模型
-        weights = train_logreg(X_train, y_train, learning_rate, iterations, batch_size, regularization, lambda_reg)
-
-        # 计算测试集准确率
-        acc = accuracy(X_test, y_test, weights)
-
-        print(f"尝试参数组合: {params}, 测试集准确率: {acc:.4f}")
-
-        if acc > best_accuracy:
-            best_accuracy = acc
-            best_params = params
-
-    return best_params, best_accuracy
 
 if __name__ == "__main__":
     X_train, y_train = load_data("./data/horse-colic.data")
     X_test, y_test = load_data("./data/horse-colic.test")
 
-    # 定义参数网格
-    param_grid = {
-        'learning_rate': [0.01, 0.05, 0.1, 0.15, 0.2, 0.25],
-        'iterations': [200, 500, 1000, 1500],
-        'batch_size': [16, 32, 64, 128],
-        'regularization': [None, 'L1', 'L2'],
-        'lambda_reg': [0.1, 0.5, 0.05, 0.15]
-    }
-
-    # 进行网格搜索
-    best_params, best_accuracy = grid_search(X_train, y_train, X_test, y_test, param_grid)
-
-    print(f"最佳参数组合: {best_params}")
-    print(f"最佳准确率: {best_accuracy:.4f}")
-
-    # 使用最佳参数组合进行最终测试
-    best_learning_rate, best_iterations, best_batch_size, best_regularization, best_lambda_reg = best_params
-    final_weights = train_logreg(X_train, y_train, best_learning_rate, best_iterations, best_batch_size, best_regularization, best_lambda_reg)
+    final_weights = train_logreg(X_train, y_train, learning_rate=0.05, iterations=1500, batch_size=64, regularization='L2', lambda_reg=0.05)
 
     # 计算并输出最终测试集的准确率
     final_accuracy = accuracy(X_test, y_test, final_weights)