使用Pandas的日期范围，每个二月有28天，我做错了什么

Question

这个数据帧的输出显示每个二月有28天

dates = pd.DataFrame({"start_date": pd.date_range(datetime.datetime(1900, 1, 1).strftime("%Y-%m-%d"), datetime.datetime(2020, 1, 1).strftime("%Y-%m-%d"), freq='MS')})
dates["start_date"] = dates["start_date"].apply(pd.datetools.normalize_date)
dates["end_date"] = dates["start_date"].shift(-1)
dates['end_date'] = dates['end_date'] - pd.offsets.Day(1)
dates["end_date"][1440]= datetime.datetime(2020, 1, 31).strftime("%Y-%m-%d")
def prettyrow(start,end):
    result = end-start+ pd.offsets.Day(1)
    return result

data = []
for i in range(0,dates.shape[0]):
    days = prettyrow(dates["start_date"][i],dates["end_date"][i])
    data.append(int(str(days)[:2]))

df = pd.DataFrame(data,columns=["days"])
timedata = pd.concat([dates, df], axis=1)
timedata

我不得不分别组合这两个数据帧，因为我无法解决如何将特定数据帧列中的每一行传递给具有多个参数的函数。你知道吗

根据下面的评论

print(timedata[(timedata.start_date.dt.year%4 == 0) & (timedata.start_date.dt.month == 2)])

     start_date   end_date  days
1    1900-02-01 1900-02-28    28
49   1904-02-01 1904-02-29    29
97   1908-02-01 1908-02-29    29
145  1912-02-01 1912-02-29    29
193  1916-02-01 1916-02-29    29
241  1920-02-01 1920-02-29    29
289  1924-02-01 1924-02-29    29
337  1928-02-01 1928-02-29    29
385  1932-02-01 1932-02-29    29
433  1936-02-01 1936-02-29    29
481  1940-02-01 1940-02-29    29
529  1944-02-01 1944-02-29    29
577  1948-02-01 1948-02-29    29
625  1952-02-01 1952-02-29    29
673  1956-02-01 1956-02-29    29
721  1960-02-01 1960-02-29    29
769  1964-02-01 1964-02-29    29
817  1968-02-01 1968-02-29    29
865  1972-02-01 1972-02-29    29
913  1976-02-01 1976-02-29    29
961  1980-02-01 1980-02-29    29
1009 1984-02-01 1984-02-29    29
1057 1988-02-01 1988-02-29    29
1105 1992-02-01 1992-02-29    29
1153 1996-02-01 1996-02-29    29
1201 2000-02-01 2000-02-29    29
1249 2004-02-01 2004-02-29    29
1297 2008-02-01 2008-02-29    29
1345 2012-02-01 2012-02-29    29
1393 2016-02-01 2016-02-29    29

以下注释的输出为：

     start_date   end_date  days
1    1900-02-01 1900-02-28    28
49   1904-02-01 1904-02-29    29
97   1908-02-01 1908-02-29    29
145  1912-02-01 1912-02-29    29
193  1916-02-01 1916-02-29    29
241  1920-02-01 1920-02-29    29
289  1924-02-01 1924-02-29    29
337  1928-02-01 1928-02-29    29
385  1932-02-01 1932-02-29    29
433  1936-02-01 1936-02-29    29
481  1940-02-01 1940-02-29    29
529  1944-02-01 1944-02-29    29
577  1948-02-01 1948-02-29    29
625  1952-02-01 1952-02-29    29
673  1956-02-01 1956-02-29    29
721  1960-02-01 1960-02-29    29
769  1964-02-01 1964-02-29    29
817  1968-02-01 1968-02-29    29
865  1972-02-01 1972-02-29    29
913  1976-02-01 1976-02-29    29
961  1980-02-01 1980-02-29    29
1009 1984-02-01 1984-02-29    29
1057 1988-02-01 1988-02-29    29
1105 1992-02-01 1992-02-29    29
1153 1996-02-01 1996-02-29    29
1201 2000-02-01 2000-02-29    29
1249 2004-02-01 2004-02-29    29
1297 2008-02-01 2008-02-29    29
1345 2012-02-01 2012-02-29    29
1393 2016-02-01 2016-02-29    29