为什么数据帧和系列之间的运算符“*”给我一个带有 NaN 的 len(firstSeries)*len(secondSeries)？
Why the operator "*" between a dataframe and a series is giving me a len(firstSeries)*len(secondSeries) with NaNs?
我得到 res = dataframe * dataframe['column'] 第一个数据帧 <class 'pandas.core.frame.DataFrame'> 和 <class 'pandas.core.series.Series'> dataframe['column']
它输出我： res : 2020-08-17 05:00:00 2020-08-17 05:15:00 2020-08-17 05:30:00 ... dtime 2020-08-24 16:45:00 NaN NaN NaN ... 2020-08-24 16:30:00 NaN NaN NaN ... 2020-08-24 16:15:00 NaN NaN NaN ... 2020-08-24 16:00:00 NaN NaN NaN ... 2020-08-24 15:45:00 NaN NaN NaN ... ... 2020-08-17 06:00:00 NaN NaN NaN ... 2020-08-17 05:45:00 NaN NaN NaN ... 2020-08-17 05:30:00 NaN NaN NaN ... 2020-08-17 05:15:00 NaN NaN NaN ... 2020-08-17 05:00:00 NaN NaN NaN ...
我发现的一切都告诉我“*”应该给我：res : dtime 2020-08-24 16:45:00 floatValue 2020-08-24 16:30:00 floatValue 2020-08-24 16:15:00 floatValue 2020-08-24 16:00:00 floatValue 2020-08-24 15:45:00 floatValue ... 2020-08-17 06:00:00 floatValue 2020-08-17 05:45:00 floatValue 2020-08-17 05:30:00 floatValue 2020-08-17 05:15:00 floatValue 2020-08-17 05:00:00 floatValue
我确定我遗漏了什么，但我找不到什么。
编辑：
import pandas
df = pandas.DataFrame({"angles": [0, 3, 4], "degrees": [360, 180, 360]}, index=["circle", "triangle", "rectangle"])
other = pandas.DataFrame({"angles": [0, 3, 4]}, index=["circle", "triangle", "rectangle"])
print("df * other :", df * other)
给我：
df * other :
angles degrees
circle 0 NaN
triangle 9 NaN
rectangle 16 NaN
符合预期。所以也许它与 datetimeIndex 有关。
编辑：
import pandas
df = pandas.DataFrame({'angles': [0, 3, 4], 'degrees': [360, 180, 360]}, index=['2020-08-17 05:30:00', '2020-08-17 05:15:00', '2020-08-17 05:00:00'])
other = pandas.DataFrame({'angles': [0, 3, 4]}, index=['2020-08-17 05:30:00', '2020-08-17 05:15:00', '2020-08-17 05:00:00'])
df.index = pandas.to_datetime(df.index, format='%Y-%m-%d %H:%M:%S')
other.index = pandas.to_datetime(other.index, format='%Y-%m-%d %H:%M:%S')
print("df * other :", df * other)
正在给予：
df * other : angles degrees
2020-08-17 05:30:00 0 NaN
2020-08-17 05:15:00 9 NaN
2020-08-17 05:00:00 16 NaN
所以这不是 datetimeIndex。
在你的小例子中，你乘以 df * df[即其他 ]。如果你使用 df *df.angles [即df * 系列]
但在主要问题中，您使用的是 df * 系列。尝试使用 df * pd.DataFrame(df.col) .
在 df * 系列中：它按列比较与 df.multiply(df.series,axis=1) 相同
你也可以试试 df.multiply(df.series,axis=0)
我得到 res = dataframe * dataframe['column'] 第一个数据帧 <class 'pandas.core.frame.DataFrame'> 和 <class 'pandas.core.series.Series'> dataframe['column']
它输出我： res : 2020-08-17 05:00:00 2020-08-17 05:15:00 2020-08-17 05:30:00 ... dtime 2020-08-24 16:45:00 NaN NaN NaN ... 2020-08-24 16:30:00 NaN NaN NaN ... 2020-08-24 16:15:00 NaN NaN NaN ... 2020-08-24 16:00:00 NaN NaN NaN ... 2020-08-24 15:45:00 NaN NaN NaN ... ... 2020-08-17 06:00:00 NaN NaN NaN ... 2020-08-17 05:45:00 NaN NaN NaN ... 2020-08-17 05:30:00 NaN NaN NaN ... 2020-08-17 05:15:00 NaN NaN NaN ... 2020-08-17 05:00:00 NaN NaN NaN ...
我发现的一切都告诉我“*”应该给我：res : dtime 2020-08-24 16:45:00 floatValue 2020-08-24 16:30:00 floatValue 2020-08-24 16:15:00 floatValue 2020-08-24 16:00:00 floatValue 2020-08-24 15:45:00 floatValue ... 2020-08-17 06:00:00 floatValue 2020-08-17 05:45:00 floatValue 2020-08-17 05:30:00 floatValue 2020-08-17 05:15:00 floatValue 2020-08-17 05:00:00 floatValue
我确定我遗漏了什么，但我找不到什么。
编辑：
import pandas
df = pandas.DataFrame({"angles": [0, 3, 4], "degrees": [360, 180, 360]}, index=["circle", "triangle", "rectangle"])
other = pandas.DataFrame({"angles": [0, 3, 4]}, index=["circle", "triangle", "rectangle"])
print("df * other :", df * other)
给我：
df * other :
angles degrees
circle 0 NaN
triangle 9 NaN
rectangle 16 NaN
符合预期。所以也许它与 datetimeIndex 有关。
编辑：
import pandas
df = pandas.DataFrame({'angles': [0, 3, 4], 'degrees': [360, 180, 360]}, index=['2020-08-17 05:30:00', '2020-08-17 05:15:00', '2020-08-17 05:00:00'])
other = pandas.DataFrame({'angles': [0, 3, 4]}, index=['2020-08-17 05:30:00', '2020-08-17 05:15:00', '2020-08-17 05:00:00'])
df.index = pandas.to_datetime(df.index, format='%Y-%m-%d %H:%M:%S')
other.index = pandas.to_datetime(other.index, format='%Y-%m-%d %H:%M:%S')
print("df * other :", df * other)
正在给予：
df * other : angles degrees
2020-08-17 05:30:00 0 NaN
2020-08-17 05:15:00 9 NaN
2020-08-17 05:00:00 16 NaN
所以这不是 datetimeIndex。
在你的小例子中，你乘以 df * df[即其他 ]。如果你使用 df *df.angles [即df * 系列]
但在主要问题中，您使用的是 df * 系列。尝试使用 df * pd.DataFrame(df.col) .
在 df * 系列中：它按列比较与 df.multiply(df.series,axis=1) 相同 你也可以试试 df.multiply(df.series,axis=0)