dataframe 采样后，再计算和报错 咋回事？

[色郎中]

df=pd.DataFrame(np.random.randn(5,5))
print(df)
print(df.apply(lambda x: x.sum()))
df1=df.loc[0].sample(3)
print(df1)
print(df1.apply(lambda x: x.sum()))

          0         1         2         3         4
0 -2.021641  0.202247 -0.357294 -0.308477  0.301995
1  1.395682  1.627884 -0.156677  0.768417 -0.348573
2  1.143972  1.952319 -0.797111  1.719115  0.636497
3  0.819878 -0.756803 -0.926412  0.177137 -0.354646
4 -0.640460  0.181148  1.037831 -0.864291  0.145361
上面是生成的数据

0    0.697431
1    3.206794
2   -1.199663
3    1.491900
4    0.380635
dtype: float64
上面是每列 求和的结果

Traceback (most recent call last):
  File "E:/2017/Anaconda3/pandasExamples/test.py", line 15, in <module>
    print(df1.apply(lambda x: x.sum()))
  File "C:\Users\sin\Anaconda3\lib\site-packages\pandas\core\series.py", line 2294, in apply
    mapped = lib.map_infer(values, f, convert=convert_dtype)
  File "pandas\src\inference.pyx", line 1207, in pandas.lib.map_infer (pandas\lib.c:66124)
  File "E:/2017/Anaconda3/pandasExamples/test.py", line 15, in <lambda>
    print(df1.apply(lambda x: x.sum()))
AttributeError: 'float' object has no attribute 'sum'
4    0.301995
3   -0.308477
0   -2.021641
Name: 0, dtype: float64
这是在第一行里随机取的3个数据，想对这3个数据求和时，报错了，这是什么情况呢？

[色郎中]

上面打印结果 对原始的数据的每列求和正常，，而采样后的数据再求和就报错 不解啊

[色郎中]

df1=pd.DataFrame(df.loc[0].sample(3))

再来一次，好像就对了