我的数据如下:
+----+------+-------+
| ID | YEAR | MONTH |
+----+------+-------+
| A | 2017 | 1 |
| B | 2017 | 2 |
| C | 2017 | 3 |
| D | 2017 | 4 |
| E | 2017 | 5 |
| F | 2017 | 6 |
+----+------+-------+
我想添加一个名为DATE
的新列,该列存储由YEAR
和MONTH
列的日期对象组成的新列。像这样:
+----+------+-------+------------+
| ID | YEAR | MONTH | DATE |
+----+------+-------+------------+
| A | 2017 | 1 | 2017-01-01 |
| B | 2017 | 2 | 2017-02-01 |
| C | 2017 | 3 | 2017-03-01 |
| D | 2017 | 4 | 2017-04-01 |
| E | 2017 | 5 | 2017-05-01 |
| F | 2017 | 6 | 2017-06-01 |
+----+------+-------+------------+
我使用以下代码创建列,但想知道是否有更清洁的'Pythonic'单行。类似df['DATE']=date(df.year, df.month, 1)
。
import pandas as pd
from datetime import date
ID = ['A', 'B', 'C', 'D', 'E', 'F']
YEAR = [2017, 2017, 2017, 2017, 2017, 2017]
MONTH = [1, 2, 3, 4, 5, 6]
df = pd.DataFrame({'ID': ID, 'YEAR': YEAR, 'MONTH': MONTH})
DATE = []
for y, m in zip(df.YEAR, df.MONTH):
DATE.append(date(y, m, 1))
df['DATE'] = DATE
答案 0 :(得分:2)
选项1
将包含3列的数据框切片 - YEAR
,MONTH
和DAY
传递给pd.to_datetime
。
df['DATE'] = pd.to_datetime(df[['YEAR', 'MONTH']].assign(DAY=1))
df
ID MONTH YEAR DATE
0 A 1 2017 2017-01-01
1 B 2 2017 2017-02-01
2 C 3 2017 2017-03-01
3 D 4 2017 2017-04-01
4 E 5 2017 2017-05-01
5 F 6 2017 2017-06-01
选项2
字符串连接,pd.to_datetime
。
pd.to_datetime(df.YEAR.astype(str) + '/' + df.MONTH.astype(str) + '/01')
0 2017-01-01
1 2017-02-01
2 2017-03-01
3 2017-04-01
4 2017-05-01
5 2017-06-01
dtype: datetime64[ns]