如何连接多个数据帧(reduce函数)，将列重命名为原始数据帧名称

Question

我有12个数据帧(july_df，aug_df……june_df(

它们都共享相同的列名，我想将它们合并到第一列，"；MINISTRY"；，并使用df的名称重命名所有列。

例如，输出可能看起来像：

Ministry item_jul item_aug.......item_jun item2_jul item2_aug.....item2jun
xyz          1       10            12        11        22           11
abc

Answer 1

你可以试试这个：

import pandas as pd
def get_df_name(df):
name = [x for x in globals() if globals()[x] is df][0]
return name
# Toy dataframes
june = pd.DataFrame(
{
"ministry": ["abc", "def", "ghi", "jkl"],
"item": [1, 2, 3, 4],
"item2": [5, 6, 7, 8],
"item3": [9, 10, 11, 12],
}
)
july = pd.DataFrame(
{
"ministry": ["abc", "def", "ghi", "jkl"],
"item": [13, 14, 15, 16],
"item2": [17, 18, 19, 20],
"item3": [21, 22, 23, 24],
}
)
august = pd.DataFrame(
{
"ministry": ["abc", "def", "ghi", "jkl"],
"item": [25, 26, 27, 28],
"item2": [29, 30, 31, 32],
"item3": [33, 34, 35, 36],
}
)
dfs = [june, july, august]
# Merge dataframes on "ministry" after renaming columns
merged_dfs = dfs[0]
merged_dfs.columns = [
f"{col}_{get_df_name(merged_dfs)}" if col != "ministry" else col
for col in merged_dfs.columns
]
for df in dfs[1:]:
df.columns = [
f"{col}_{get_df_name(df)}" if col != "ministry" else col for col in df.columns
]
merged_dfs = merged_dfs.merge(df, on="ministry")
print(merged_dfs)
# Outputs
ministry  item_june  item2_june  ...  item_august  item2_august  item3_august
0      abc          1           5  ...           25            29            33
1      def          2           6  ...           26            30            34
2      ghi          3           7  ...           27            31            35
3      jkl          4           8  ...           28            32            36