更新 function.py

function.py

@@ -21,16 +21,17 @@ def load_data(filepath):
     - np.array: 标签数据。
     """
     data = np.genfromtxt(filepath, delimiter=' ', dtype=float, missing_values='?', filling_values=np.nan)
-    nan_columns = np.isnan(data)
+    # 填充缺失值为列均值
     col_means = np.nanmean(data, axis=0)
-    data[nan_columns] = np.take(col_means, np.where(nan_columns)[1])
+    inds = np.where(np.isnan(data))
+    data[inds] = np.take(col_means, inds[1])
     
     X = data[:, :-1]
     y = data[:, -1]
     return X, y
 
 # 对率回归训练函数
-def train_logreg(X, y, learning_rate=0.01, iterations=1000):
+def train_logreg(X, y, learning_rate=0.01, iterations=500):
     """
     对率回归训练函数。
 
@@ -47,6 +48,7 @@ def train_logreg(X, y, learning_rate=0.01, iterations=1000):
     X = np.c_[X, np.ones(n_samples)]  # 添加偏置项
     weights = np.zeros(n_features + 1)
 
+    # 向量化梯度下降
     for _ in range(iterations):
         y_pred = sigmoid(np.dot(X, weights))
         gradient = np.dot(X.T, (y_pred - y)) / n_samples
@@ -66,8 +68,6 @@ def predict_logreg(X, weights):
     Returns:
     - np.array: 预测的分类标签 (0 或 1)。
     """
-    if X.ndim == 1:
-        X = X.reshape(1, -1)
     X = np.c_[X, np.ones(X.shape[0])]
     y_pred = sigmoid(np.dot(X, weights))
     return (y_pred >= 0.5).astype(int)
@@ -78,7 +78,7 @@ if __name__ == "__main__":
     X_test, y_test = load_data("./data/horse-colic.test")
 
     # 训练对率回归模型
-    weights = train_logreg(X_train, y_train, learning_rate=0.01, iterations=1000)
+    weights = train_logreg(X_train, y_train, learning_rate=0.01, iterations=500)
 
     # 预测测试集并计算准确率
     predictions = predict_logreg(X_test, weights)