对于此数据帧
df = pd.DataFrame({'B': [0, 1, 2, np.nan, 4]},index=[0.1,0.2,0.5,0.7,0.9])
df
B
0.1 0.0
0.2 1.0
0.5 2.0
0.7 NaN
0.9 4.0
我想执行基于偏移的滚动,如:
df.rolling(0.3).sum()
但我得到了这个错误:
ValueError:窗口必须是整数
有人能帮我吗?我的真实案例涉及大约1B行,最后我需要执行groupby((.rolling((。以上只是对我的问题的极端简化。请帮帮我。谢谢!
最好的答案感谢GitHub中的mroeschke:
In [1]: df = pd.DataFrame({'B': [0, 1, 2, np.nan, 4]},index=[0.1,0.2,0.5,0.7,0.9])
In [2]: df.index = pd.to_datetime(df.index, unit="s")
In [3]: df.rolling("300ms").mean()
Out[3]:
B
1970-01-01 00:00:00.100 0.0
1970-01-01 00:00:00.200 0.5
1970-01-01 00:00:00.500 2.0
1970-01-01 00:00:00.700 2.0
1970-01-01 00:00:00.900 4.0
最好的答案感谢GitHub中的mroeschke:
In [1]: df = pd.DataFrame({'B': [0, 1, 2, np.nan, 4]},index=[0.1,0.2,0.5,0.7,0.9])
In [2]: df.index = pd.to_datetime(df.index, unit="s")
In [3]: df.rolling("300ms").mean()
Out[3]:
B
1970-01-01 00:00:00.100 0.0
1970-01-01 00:00:00.200 0.5
1970-01-01 00:00:00.500 2.0
1970-01-01 00:00:00.700 2.0
1970-01-01 00:00:00.900 4.0