我有一个日期格式为%Y-%m-%d %H:%M:%S或%Y-%m-%d的数据框：
...
87986 1979-06-18 00:00:00
87987 1979-06-18 00:00:00
87988 1987-03-18
87989 1983-11-01
...
我想用相同的方式格式化它们。我尝试过：
df['birthdate']=pd.to_datetime(df['birthdate'].astype(str), format='%Y-%m-%d')
但是回来了：
---------------------------------------------------------------------------
TypeError Traceback (most recent call last)
C:\ProgramData\Anaconda3\lib\site-packages\pandas\core\arrays\datetimes.py in objects_to_datetime64ns(data, dayfirst, yearfirst, utc, errors, require_iso8601, allow_object)
1860 try:
-> 1861 values, tz_parsed = conversion.datetime_to_datetime64(data)
1862 # If tzaware, these values represent unix timestamps, so we
pandas/_libs/tslibs/conversion.pyx in pandas._libs.tslibs.conversion.datetime_to_datetime64()
TypeError: Unrecognized value type: <class 'str'>
During handling of the above exception, another exception occurred:
ValueError Traceback (most recent call last)
<ipython-input-345-d5c1036edc26> in <module>
10 return born
11
---> 12 df['birthdate']=pd.to_datetime(df['birthdate'].astype(str), format='%Y-%m-%d')
13 df["age"] = df["birthdate"].apply(calculate_age)
C:\ProgramData\Anaconda3\lib\site-packages\pandas\core\tools\datetimes.py in to_datetime(arg, errors, dayfirst, yearfirst, utc, box, format, exact, unit, infer_datetime_format, origin, cache)
590 else:
591 from pandas import Series
--> 592 values = convert_listlike(arg._values, True, format)
593 result = Series(values, index=arg.index, name=arg.name)
594 elif isinstance(arg, (ABCDataFrame, compat.MutableMapping)):
C:\ProgramData\Anaconda3\lib\site-packages\pandas\core\tools\datetimes.py in _convert_listlike_datetimes(arg, box, format, name, tz, unit, errors, infer_datetime_format, dayfirst, yearfirst, exact)
300 arg, dayfirst=dayfirst, yearfirst=yearfirst,
301 utc=utc, errors=errors, require_iso8601=require_iso8601,
--> 302 allow_object=True)
303
304 if tz_parsed is not None:
C:\ProgramData\Anaconda3\lib\site-packages\pandas\core\arrays\datetimes.py in objects_to_datetime64ns(data, dayfirst, yearfirst, utc, errors, require_iso8601, allow_object)
1864 return values.view('i8'), tz_parsed
1865 except (ValueError, TypeError):
-> 1866 raise e
1867
1868 if tz_parsed is not None:
C:\ProgramData\Anaconda3\lib\site-packages\pandas\core\arrays\datetimes.py in objects_to_datetime64ns(data, dayfirst, yearfirst, utc, errors, require_iso8601, allow_object)
1855 dayfirst=dayfirst,
1856 yearfirst=yearfirst,
-> 1857 require_iso8601=require_iso8601
1858 )
1859 except ValueError as e:
pandas/_libs/tslib.pyx in pandas._libs.tslib.array_to_datetime()
pandas/_libs/tslib.pyx in pandas._libs.tslib.array_to_datetime()
ValueError: time data None doesn't match format specified
但是，我的行都不是None。
然后我尝试：
df['birthdate']=pd.to_datetime(df['birthdate']).dt.strftime('%Y-%m-%d')
但是得到了：
---------------------------------------------------------------------------
TypeError Traceback (most recent call last)
C:\ProgramData\Anaconda3\lib\site-packages\pandas\core\arrays\datetimes.py in objects_to_datetime64ns(data, dayfirst, yearfirst, utc, errors, require_iso8601, allow_object)
1860 try:
-> 1861 values, tz_parsed = conversion.datetime_to_datetime64(data)
1862 # If tzaware, these values represent unix timestamps, so we
pandas/_libs/tslibs/conversion.pyx in pandas._libs.tslibs.conversion.datetime_to_datetime64()
TypeError: Unrecognized value type: <class 'str'>
During handling of the above exception, another exception occurred:
OutOfBoundsDatetime Traceback (most recent call last)
<ipython-input-350-b23ba7455ab9> in <module>
----> 1 df['birthdate']=pd.to_datetime(df['birthdate']).dt.strftime('%Y-%m-%d')
C:\ProgramData\Anaconda3\lib\site-packages\pandas\core\tools\datetimes.py in to_datetime(arg, errors, dayfirst, yearfirst, utc, box, format, exact, unit, infer_datetime_format, origin, cache)
590 else:
591 from pandas import Series
--> 592 values = convert_listlike(arg._values, True, format)
593 result = Series(values, index=arg.index, name=arg.name)
594 elif isinstance(arg, (ABCDataFrame, compat.MutableMapping)):
C:\ProgramData\Anaconda3\lib\site-packages\pandas\core\tools\datetimes.py in _convert_listlike_datetimes(arg, box, format, name, tz, unit, errors, infer_datetime_format, dayfirst, yearfirst, exact)
300 arg, dayfirst=dayfirst, yearfirst=yearfirst,
301 utc=utc, errors=errors, require_iso8601=require_iso8601,
--> 302 allow_object=True)
303
304 if tz_parsed is not None:
C:\ProgramData\Anaconda3\lib\site-packages\pandas\core\arrays\datetimes.py in objects_to_datetime64ns(data, dayfirst, yearfirst, utc, errors, require_iso8601, allow_object)
1864 return values.view('i8'), tz_parsed
1865 except (ValueError, TypeError):
-> 1866 raise e
1867
1868 if tz_parsed is not None:
C:\ProgramData\Anaconda3\lib\site-packages\pandas\core\arrays\datetimes.py in objects_to_datetime64ns(data, dayfirst, yearfirst, utc, errors, require_iso8601, allow_object)
1855 dayfirst=dayfirst,
1856 yearfirst=yearfirst,
-> 1857 require_iso8601=require_iso8601
1858 )
1859 except ValueError as e:
pandas/_libs/tslib.pyx in pandas._libs.tslib.array_to_datetime()
pandas/_libs/tslib.pyx in pandas._libs.tslib.array_to_datetime()
pandas/_libs/tslib.pyx in pandas._libs.tslib.array_to_datetime()
pandas/_libs/tslib.pyx in pandas._libs.tslib.array_to_datetime()
pandas/_libs/tslibs/np_datetime.pyx in pandas._libs.tslibs.np_datetime.check_dts_bounds()
OutOfBoundsDatetime: Out of bounds nanosecond timestamp: 7974-04-23 00:00:00
calculate_age函数这种格式的全部原因是为了能够计算出出生日期到现在之间的年龄/时间。我建立的功能是：
from datetime import datetime, date
def calculate_age(born):
today = date.today
days_in_year = 365.2425
if born not in [None, 'NaT']:
age = int((date.today() - born.date()).days / days_in_year)
return age
else:
return born
df['birthdate']=pd.to_datetime(df['birthdate'], errors='coerce')
df["age"] = df["birthdate"].apply(calculate_age)
答案 0 :(得分：1)
我认为没有必要转换为字符串，如果两种不同的格式也不能指定格式，但是对于将不可解析的日期时间（如errors='coerce'转换为缺失值7974-04-23 00:00:00，NaT是必需的：
print (df)
birthdate
87986 1979-06-18 00:00:00
87987 1979-06-18 00:00:00
87988 1987-03-18
87989 1983-11-01
87990 7974-04-23 00:00:00 <- added row
df['birthdate']=pd.to_datetime(df['birthdate'], errors='coerce')
print (df)
birthdate
87986 1979-06-18
87987 1979-06-18
87988 1987-03-18
87989 1983-11-01
87990 NaT
然后最简单的是删除缺失值：
df["age"] = df["birthdate"].dropna().apply(calculate_age)
更改后的功能有什么
from datetime import datetime, date
def calculate_age(born):
today = date.today
days_in_year = 365.2425
if pd.notna(born):
age = int((date.today() - born.date()).days / days_in_year)
return age
else:
return born
df["age"] = df["birthdate"].apply(calculate_age)
print (df)
birthdate age
87986 1979-06-18 40.0
87987 1979-06-18 40.0
87988 1987-03-18 32.0
87989 1983-11-01 36.0
87990 NaT NaN