多索引pandas数据帧中的过滤

import pandas as pd

# 创建多索引
arrays = [
    ['2020', '2020', '2021', '2021'],
    ['Jan', 'Feb', 'Jan', 'Feb']
]
index = pd.MultiIndex.from_arrays(arrays, names=('Year', 'Month'))

# 创建 DataFrame
data = {'Sales': [100, 150, 200, 250]}
df = pd.DataFrame(data, index=index)

print(df)

              Sales
Year Month       
2020 Jan      100
      Feb      150
2021 Jan      200
      Feb      250

# 使用 .loc 过滤
filtered_df = df.loc[2021]
print(filtered_df)

# 使用 .xs 过滤
filtered_df = df.xs(2021, level='Year')
print(filtered_df)

         Sales
Month         
Jan      200
Feb      250

         Sales
Month         
Jan      200
Feb      250

# 假设我们有一个包含重复值的多索引 DataFrame
arrays = [
    ['2020', '2020', '2021', '2021'],
    ['Jan', 'Feb', 'Jan', 'Feb']
]
index = pd.MultiIndex.from_arrays(arrays, names=('Year', 'Month'))

data = {'Sales': [100, 150, 200, 250]}
df = pd.DataFrame(data, index=index)

# 去除重复值
df = df[~df.index.duplicated(keep='first')]

# 过滤
filtered_df = df.loc[2021]
print(filtered_df)

         Sales
Month         
Jan      200
Feb      250