更新 function.py

function.py

@@ -81,18 +81,39 @@ def load_data(filepath):
     """
     data = []
     labels = []
+
+    # 读取数据并处理缺失值
     with open(filepath, 'r') as file:
         for line in file:
             fields = line.strip().split()
-            if '?' in fields or len(fields) < 2:
-                continue
-            try:
-                data.append([float(x) for x in fields[:-1]])
-                labels.append(int(fields[-1]))
-            except ValueError:
+            if len(fields) < 2:
                 continue
             
-    return np.array(data), np.array(labels)
+            features = fields[:-1]
+            label = fields[-1]
+            
+            # 记录特征和标签
+            data.append(features)
+            labels.append(label)
+    
+    # 转换数据为 NumPy 数组，方便进一步处理
+    data = np.array(data)
+    labels = np.array(labels, dtype=int)
+
+    # 用于存储浮点型数据
+    numeric_data = np.zeros(data.shape, dtype=float)
+
+    # 对每列进行缺失值填补
+    for col in range(data.shape[1]):
+        # 获取列数据，将 "?" 替换为 NaN，然后计算均值
+        col_data = np.where(data[:, col] == '?', np.nan, data[:, col]).astype(float)
+        mean_value = np.nanmean(col_data)  # 计算均值忽略 NaN
+
+        # 将 NaN 替换为均值
+        col_data = np.where(np.isnan(col_data), mean_value, col_data)
+        numeric_data[:, col] = col_data
+
+    return numeric_data, labels
 
 
 if __name__ == "__main__":