如何获得 Timeindex，其中两个 Dataframes 在 Row 中都只有 NaN？答案

【问题标题】：How get Timeindex, where Two Dataframes both have only NaNs in Row?如何获得 Timeindex，其中两个 Dataframes 在 Row 中都只有 NaN？
【发布时间】：2020-12-24 15:44:34
【问题描述】：

我有两个数据帧 X 和 Y（都有一个时间索引）。两者都有一个交集，但也有另一个不一定包含的索引。

如何获取时间索引，相交和两者row is only-NaNs？

可重现：

import numpy as np, pandas as pd
X = pd.DataFrame( {"a":[1,np.nan,3,4,5,np.nan,7,8,100,9,np.nan,np.nan,12,13,14,15],"b":[1,np.nan,3,4,5,6,7,8,101,9,np.nan,np.nan,12,13,np.nan,15]}, 
                    index =pd.DatetimeIndex(["2019-07-18 08:51:00", "2019-07-18 08:52:00","2019-07-18 08:53:00","2019-07-18 08:54:00","2019-07-18 08:55:00","2019-07-18 08:56:00","2019-07-18 08:57:00","2019-07-18 08:58:00","2019-07-18 08:58:30","2019-07-18 08:59:00","2019-07-18 09:00:00","2019-07-18 09:01:00","2019-07-18 09:02:00","2019-07-18 09:03:00","2019-07-18 09:04:00","2019-07-18 09:05:00" ]))
Y = pd.DataFrame({"c":[0,1,np.nan,3,4,5,6,7,8,9,np.nan,np.nan,12,13,14,15,16],"d":[0,1,np.nan,3,4,5,6,7,8,9,np.nan,np.nan,12,13,14,np.nan,16]}, 
                    index =pd.DatetimeIndex(["2019-07-18 08:50:00","2019-07-18 08:51:00", "2019-07-18 08:52:00","2019-07-18 08:53:00","2019-07-18 08:54:00","2019-07-18 08:55:00","2019-07-18 08:56:00","2019-07-18 08:57:00","2019-07-18 08:58:00","2019-07-18 08:59:00","2019-07-18 09:00:00","2019-07-18 09:01:00","2019-07-18 09:02:00","2019-07-18 09:03:00","2019-07-18 09:04:00","2019-07-18 09:05:00","2019-07-18 09:06:00" ]))

# expected result: pd.DatetimeIndex(['2019-07-18 08:52:00', '2019-07-18 09:00:00', '2019-07-18 09:01:00'])

编辑：这有效：

intersection_X_Y = X.index.intersection(Y.index)  
result  = X.loc[intersection_X_Y].isnull().all(1) & Y.loc[intersection_X_Y].isnull().all(1)
print("result",result [result ].index)

【问题讨论】：

标签： python pandas numpy dataframe indexing

【解决方案1】：

试试这个

i_X = X.index[X.join(Y).isna().all(1)]

Out[20]:
DatetimeIndex(['2019-07-18 08:52:00', '2019-07-18 09:00:00',
               '2019-07-18 09:01:00'],
              dtype='datetime64[ns]', freq=None)

【讨论】：