pandas上的一个Timedelta的例子

该例子出自pandas-for-everyone一书.
使用了如下的csv文件.

In [80]: !cat ./gapminder/other_csv/scientists.csv
Name,Born,Died,Age,Occupation
Rosaline Franklin,1920-07-25,1958-04-16,37,Chemist
William Gosset,1876-06-13,1937-10-16,61,Statistician
Florence Nightingale,1820-05-12,1910-08-13,90,Nurse
Marie Curie,1867-11-07,1934-07-04,66,Chemist
Rachel Carson,1907-05-27,1964-04-14,56,Biologist
John Snow,1813-03-15,1858-06-16,45,Physician
Alan Turing,1912-06-23,1954-06-07,41,Computer Scientist
Johann Gauss,1777-04-30,1855-02-23,77,Mathematician

读入csv文件,将其中的一列赋值给ages,然后shuffle这一列,会发现ages会跟着变化，所以ages只是指向这一列的指针而已。

In [81]: import numpy as np

In [82]: import pandas as pd

In [83]: import matplotlib.pyplot as plt

In [84]: scientists=pd.read_csv("./gapminder/other_csv/scientists.csv")

In [85]: scientists.shape
Out[85]: (8, 5)

In [86]: scientists.columns
Out[86]: Index(['Name', 'Born', 'Died', 'Age', 'Occupation'], dtype='object')

In [87]: scientists.dtypes
Out[87]:
Name object
Born          object
Died          object
Age            int64
Occupation    object
dtype: object

In [88]: ages=scientists['Age']

In [89]: ages
Out[89]:
0    37
1    61
2    90
3    66
4    56
5    45
6    41
7    77
Name: Age, dtype: int64

In [90]: import random

In [91]: random.seed(42)

In [92]: random.shuffle(scientists['Age'])
/usr/lib/python3.7/random.py:278: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame

See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy
  x[i], x[j] = x[j], x[i]

然后根据born和died两列创造出新的两列,　类型为datetime64.

In [93]: born_datetime=pd.to_datetime(scientists['Born'], format="%Y-%m-%d")

In [94]: died_datetime=pd.to_datetime(scientists['Died'], format='%Y-%m-%d')

In [95]: scientists['Born_dt'], scientists['Died_dt']=(born_datetime, died_datetime)

In [96]: scientists.dtypes
Out[96]:
Name object
Born                  object
Died                  object
Age                    int64
Occupation            object
Born_dt       datetime64[ns]
Died_dt       datetime64[ns]
dtype: object

In [97]: scientists.shape
Out[97]: (8, 7)

最后使用得到的两列datetime64的类型做减法，得到timedelta64数据类型，然后将这个类型转化为int.

#下面两种方法都可以
scientists['age_years_dt']=scientists['age_days_dt'].astype(pd.Timedelta).apply(lambda l: l.days //365)
scientists['age_years_dt']=scientists['age_days_dt'].astype('timedelta64[D]').astype(int) // 365
In [102]: scientists.dtypes
Out[102]: 
Name                     object
Born                     object
Died                     object
Age                       int64
Occupation               object
Born_dt          datetime64[ns]
Died_dt          datetime64[ns]
age_days_dt     timedelta64[ns]
age_years_dt              int64
dtype: object

In [103]: scientists['age_years_dt']
Out[103]: 
0    37
1    61
2    90
3    66
4    56
5    45
6    41
7    77
Name: age_years_dt, dtype: int64

a