如何在Pandas中对多列应用类型转换且保留Id列？

我有一个包含Id列和多个整数列的DataFrame，希望将这些整数列转换为categorical类型，但保持Id列不变。目前的方法会丢弃Id列，请问如何实现该转换且不丢弃Id列？

当前代码：

df= df.loc[:, df.columns != 'Id'].apply(lambda x: x.astype('category'))

示例DataFrame：

{'Id': {0: 0, 1: 1, 2: 2, 3: 3, 4: 4}, 'Foundation': {0: 2, 1: 1, 2: 2, 3: 0, 4: 2}, 'GarageFinish': {0: 1, 1: 1, 2: 1, 3: 2, 4: 1}, 'LandSlope': {0: 0, 1: 0, 2: 0, 3: 0, 4: 0}, 'LotConfig': {0: 4, 1: 2, 2: 4, 3: 0, 4: 2}, 'GarageQual': {0: 4, 1: 4, 2: 4, 3: 4, 4: 4}, 'GarageCond': {0: 4, 1: 4, 2: 4, 3: 4, 4: 4}, 'LandContour': {0: 3, 1: 3, 2: 3, 3: 3, 4: 3}, 'Utilities': {0: 0, 1: 0, 2: 0, 3: 0, 4: 0}, 'GarageType': {0: 1, 1: 1, 2: 1, 3: 5, 4: 1}, 'LotShape': {0: 3, 1: 3, 2: 0, 3: 0, 4: 0}, 'Alley': {0: 2, 1: 2, 2: 2, 3: 2, 4: 2}, 'Street': {0: 1, 1: 1, 2: 1, 3: 1, 4: 1}, 'PoolQC': {0: 3, 1: 3, 2: 3, 3: 3, 4: 3}, 'Fence': {0: 4, 1: 4, 2: 4, 3: 4, 4: 4}, 'MiscFeature': {0: 4, 1: 4, 2: 4, 3: 4, 4: 4}, 'MSZoning': {0: 3, 1: 3, 2: 3, 3: 3, 4: 3}, 'SaleType': {0: 8, 1: 8, 2: 8, 3: 8, 4: 8}, 'PavedDrive': {0: 2, 1: 2, 2: 2, 3: 2, 4: 2}, 'FireplaceQu': {0: 5, 1: 4, 2: 4, 3: 2, 4: 4}, 'Condition1': {0: 2, 1: 1, 2: 2, 3: 2, 4: 2}, 'Functional': {0: 6, 1: 6, 2: 6, 3: 6, 4: 6}, 'BsmtQual': {0: 2, 1: 2, 2: 2, 3: 3, 4: 2}, 'BsmtCond': {0: 3, 1: 3, 2: 3, 3: 1, 4: 3}, 'BsmtExposure': {0: 3, 1: 1, 2: 2, 3: 3, 4: 0}, 'BsmtFinType1': {0: 2, 1: 0, 2: 2, 3: 0, 4: 2}, 'ExterQual': {0: 2, 1: 3, 2: 2, 3: 3, 4: 2}, 'BsmtFinType2': {0: 5, 1: 5, 2: 5, 3: 5, 4: 5}, 'MasVnrType': {0: 1, 1: 2, 2: 1, 3: 2, 4: 1}, 'Exterior2nd': {0: 13, 1: 8, 2: 13, 3: 15, 4: 13}, 'Heating': {0: 1, 1: 1, 2: 1, 3: 1, 4: 1}, 'Neighborhood': {0: 5, 1: 24, 2: 5, 3: 6, 4: 15}, 'SaleCondition': {0: 4, 1: 4, 2: 4, 3: 0, 4: 4}, 'Electrical': {0: 4, 1: 4, 2: 4, 3: 4, 4: 4}, 'Exterior1st': {0: 12, 1: 8, 2: 12, 3: 13, 4: 12}, 'RoofMatl': {0: 1, 1: 1, 2: 1, 3: 1, 4: 1}, 'RoofStyle': {0: 1, 1: 1, 2: 1, 3: 1, 4: 1}, 'HouseStyle': {0: 5, 1: 2, 2: 5, 3: 5, 4: 5}, 'BldgType': {0: 0, 1: 0, 2: 0, 3: 0, 4: 0}, 'Condition2': {0: 2, 1: 2, 2: 2, 3: 2, 4: 2}, 'KitchenQual': {0: 2, 1: 3, 2: 2, 3: 2, 4: 2}, 'ExterCond': {0: 4, 1: 4, 2: 4, 3: 4, 4: 4}, 'CentralAir': {0: 1, 1: 1, 2: 1, 3: 1, 4: 1}, 'HeatingQC': {0: 0, 1: 0, 2: 0, 3: 2, 4: 0}}

解决方案 #

方法1：批量修改目标列类型（推荐） #

直接定位除Id外的所有列，批量转换类型，原DataFrame结构会完整保留：

# 获取非Id列的列表
non_id_cols = df.columns[df.columns != 'Id']
# 转换这些列为categorical类型
df[non_id_cols] = df[non_id_cols].astype('category')

方法2：合并处理后的列与Id列 #

如果需要先单独处理非Id列，可以将处理结果和原Id列拼接：

import pandas as pd

# 处理非Id列
categorical_df = df.loc[:, df.columns != 'Id'].astype('category')
# 合并Id列和处理后的列
df = pd.concat([df['Id'], categorical_df], axis=1)

方法3：使用apply条件转换 #

通过apply遍历每一列，仅对非Id列进行类型转换：

df = df.apply(lambda col: col.astype('category') if col.name != 'Id' else col)

验证转换结果：执行df.dtypes可以查看各列类型，确认Id列保持原有整数类型，其他列均转为category类型。

内容的提问来源于stack exchange，提问作者Katsu