使用多索引Pandas数据帧中表达式中的其他变量创建新变量

import pandas as pd

# 创建一个多索引DataFrame
arrays = [
    ['A', 'A', 'B', 'B'],
    ['one', 'two', 'one', 'two']
]
index = pd.MultiIndex.from_arrays(arrays, names=('first', 'second'))
df = pd.DataFrame({'value': [10, 20, 30, 40]}, index=index)
print(df)

              value
first second       
A     one        10
      two        20
B     one        30
      two        40

# 假设我们要创建一个新变量 new_value，它是 value 的两倍
df['new_value'] = df['value'] * 2
print(df)

              value  new_value
first second                    
A     one        10          20
      two        20          40
B     one        30          60
      two        40          80

# 假设我们要创建一个新变量 is_large，如果 value 大于 25 则为 True，否则为 False
df['is_large'] = df['value'] > 25
print(df)

              value  new_value  is_large
first second                              
A     one        10          20     False
      two        20          40     False
B     one        30          60      True
      two        40          80      True

# 定义一个函数来处理复杂的逻辑
def custom_function(row):
    if row['value'] > 25:
        return 'High'
    else:
        return 'Low'

# 应用该函数到每一行
df['category'] = df.apply(custom_function, axis=1)
print(df)

              value  new_value  is_large category
first second                              
A     one        10          20     False      Low
      two        20          40     False      Low
B     one        30          60      True       High
      two        40          80      True       High