Question

数据框示例

    Date        Code    Count_1 Count_2
    01-01-2021  A       5       4
    04-01-2021  A       5       5
    05-01-2021  A       7       5
    07-01-2021  A       6       7
    05-02-2021  B       8       7
    09-02-2021  B       4       3
    10-02-2021  B       4       5
    11-02-2021  B       6       7
    .
    .
    .

如何添加 starting date(01-01-2021) 和 current date（按 Code 分组）之间的所有缺失值（如果有），而 {{1} 中的值} 和 Count_1 为相关缺失日期等于 NaN。

Answer 1

将 DataFrame.reindex 与唯一的 code 和 date_range 用于日期和代码对的所有组合：

df['Date'] = pd.to_datetime(df['Date'])

idx = pd.date_range('01-01-2021', pd.to_datetime('now').normalize())
codes = df['Code'].unique()

df1 = (df.set_index(['Date','Code'])
         .reindex(pd.MultiIndex.from_product([idx, codes], names=['Date','Code']))
         .reset_index())
print (df1)
          Date Code  Count_1  Count_2
0   2021-01-01    A      5.0      4.0
1   2021-01-01    B      NaN      NaN
2   2021-01-02    A      NaN      NaN
3   2021-01-02    B      NaN      NaN
4   2021-01-03    A      NaN      NaN
..         ...  ...      ...      ...
345 2021-06-22    B      NaN      NaN
346 2021-06-23    A      NaN      NaN
347 2021-06-23    B      NaN      NaN
348 2021-06-24    A      NaN      NaN
349 2021-06-24    B      NaN      NaN

[350 rows x 4 columns]

Pandas：将缺失的日期添加到当前日期

1 个答案: