更新 function.py

function.py

@@ -105,6 +105,30 @@ def accuracy(X, y, weights):
     predictions = predict_logreg(X, weights)
     return np.mean(predictions == y)
 
+def loadDataSet(filename='dataSet.txt'):
+    dataList = []
+    labelList = []
+    fr = open(filename)
+    for line in fr.readlines():
+        lineList = line.strip().split()
+        # 修改前
+        dataList.append([1.0, float(lineList[0]), float(lineList[1])])
+        # 修改后
+        # dataList.append([float(lineList[0]), float(lineList[1])])
+        labelList.append(int(lineList[2]))
+    return dataList, labelList
+
+# 假设这里有一个简单的线性回归模型训练和评估的函数
+def trainAndEvaluate(dataList, labelList):
+    # 这里只是一个简单的示例，实际中可能需要使用更复杂的模型和评估方法
+    import numpy as np
+    X = np.array(dataList)
+    y = np.array(labelList)
+    w = np.linalg.inv(X.T @ X) @ X.T @ y
+    predictions = X @ w
+    error = np.mean((predictions - y)**2)
+    return error
+
 if __name__ == "__main__":
     X_train, y_train = load_data("./horse-colic.data")
     X_test, y_test = load_data("./horse-colic.test")
@@ -121,3 +145,13 @@ if __name__ == "__main__":
     final_accuracy = accuracy(X_test, y_test, final_weights)
     print(f"最终测试集的准确率: {final_accuracy:.4f}")
 
+    # 读取原始数据
+    dataList1, labelList1 = loadDataSet()
+    error1 = trainAndEvaluate(dataList1, labelList1)
+
+    # 修改代码后读取数据
+    dataList2, labelList2 = loadDataSet()
+    error2 = trainAndEvaluate(dataList2, labelList2)
+
+    print(f"有常数项时的误差：{error1}")
+    print(f"无常数项时的误差：{error2}")