在python中合并具有不同时间频率的系列/数据帧

时间:2018-01-14 19:28:55

标签: python pandas datetime dataframe merge

我正在尝试合并两个系列,一个小时,另一个分钟(并将索引保​​持在分钟级别):

Series A:

time
2017-09-01 01:00:00   0.5
2017-09-01 02:00:00   0.4
Freq: H, Name: A, dtype: float64

系列B分钟:

Series B

time
2017-09-01 00:00:00         NaN
2017-09-01 00:03:00   -0.000350
2017-09-01 00:06:00    0.000401
Name: B, dtype: float64

我想按小时合并这两个系列。所需的合并数据框将是:

time
2017-09-01 00:00:00         NaN       0.5
2017-09-01 00:03:00   -0.000350       0.5
2017-09-01 00:06:00    0.000401       0.5
2017-09-01 01:00:00    0.002301       0.4
2017-09-01 01:03:00    0.005601       0.4

搜索SO之后,我找到的一种方法是创建一个小时' B系列中的列,然后使用A的索引和'小时'加入A到B. B栏:(参考:Merging/combining two dataframes with different frequency time series indexes in Pandas?

B = B.to_frame()
B['hour'] = B.index.to_period('H')
merged = B.join(A, on = 'hour', how = 'left')

但是我收到了错误消息:

TypeError: Argument 'values' has incorrect type (expected numpy.ndarray, got Index)

有谁知道如何解决这个问题?或者也许有更好的方式加入这两个系列?非常感谢!

1 个答案:

答案 0 :(得分:1)

演示:

In [280]: A
Out[280]:
time
2017-09-01 01:00:00    0.5
2017-09-01 02:00:00    0.4
Name: val, dtype: float64

In [281]: B
Out[281]:
time
2017-09-01 00:00:00         NaN
2017-09-01 00:03:00   -0.000350
2017-09-01 00:06:00    0.000401
Name: val, dtype: float64

In [282]: B.to_frame('B').join(A.to_frame('A').set_index(A.index.shift(-1, freq='H')).resample('3T').ffill())
Out[282]:
                            B    A
time
2017-09-01 00:00:00       NaN  0.5
2017-09-01 00:03:00 -0.000350  0.5
2017-09-01 00:06:00  0.000401  0.5
相关问题