Contents

数据对齐

import pandas as pd
import numpy as np

广播机制

DataFrame 对象之间的数据对齐在列和索引（行标签）上自动对齐。同样，结果对象将得到列和行标签的并集。

df1 = pd.DataFrame(np.random.randn(10, 4), columns=["A", "B", "C", "D"])
df1

	A	B	C	D
0	-0.205518	-1.079244	0.466754	0.697197
1	0.373488	-0.704131	-0.562608	-0.747020
2	0.727782	1.176981	1.748998	-2.381428
3	0.693188	-0.236861	-0.045300	1.116772
4	0.679781	-0.113023	-0.237897	0.991700
5	0.301815	-0.736582	-0.327876	1.240366
6	-0.362792	0.409796	-0.559092	1.486711
7	1.158763	2.898433	-1.066979	1.109399
8	1.801811	-0.150998	0.256121	0.626933
9	0.248901	-0.846965	0.728085	-0.302709

df2 = pd.DataFrame(np.random.randn(7, 3), columns=["A", "B", "C"])
df2

	A	B	C
0	0.656296	0.948401	1.203846
1	0.576805	2.193866	-1.295344
2	0.107251	-1.394675	1.322735
3	0.278274	-0.398505	-0.894721
4	0.798450	-0.817746	1.933429
5	-0.856174	0.212137	-0.323455
6	-0.148207	2.293063	0.164304

df1 + df2

	A	B	C	D
0	0.450778	-0.130843	1.670600	NaN
1	0.950293	1.489735	-1.857952	NaN
2	0.835033	-0.217694	3.071733	NaN
3	0.971461	-0.635366	-0.940020	NaN
4	1.478231	-0.930769	1.695532	NaN
5	-0.554358	-0.524445	-0.651331	NaN
6	-0.510998	2.702860	-0.394789	NaN
7	NaN	NaN	NaN	NaN
8	NaN	NaN	NaN	NaN
9	NaN	NaN	NaN	NaN

在 DataFrame 和 Series 之间执行操作时，默认行为是对齐 DataFrame 列上的 Series 索引，从而按行广播。例如：

df1 - df1.iloc[0]

	A	B	C	D
0	0.000000	0.000000	0.000000	0.000000
1	0.579007	0.375113	-1.029362	-1.444217
2	0.933301	2.256226	1.282244	-3.078625
3	0.898706	0.842383	-0.512054	0.419576
4	0.885300	0.966221	-0.704651	0.294503
5	0.507334	0.342662	-0.794630	0.543169
6	-0.157273	1.489041	-1.025846	0.789514
7	1.364281	3.977677	-1.533733	0.412202
8	2.007330	0.928246	-0.210632	-0.070263
9	0.454420	0.232279	0.261331	-0.999906

Last updated on Feb 19, 2022