如何从数组中的嵌套值列表、字典中的值组成的输出中构造熊猫数据

Question

此类变量的数据输出如下：

{'coupling.2.b': array([[0.00247262],
    [0.00248247],
    [0.00249233],
    ...,
    [0.01106487],
    [0.01106487],
    [0.01106487]]), 'coupling.2.ca': array([[9.98543405e-09],
    [9.97095234e-09],
    [9.95652127e-09],
    ...,
    [1.22356842e-08],
    [1.22356843e-08],
    [1.22356843e-08]]), 'coupling.2.gnmda': array([[0.002     ],
    [0.002     ],
    [0.002     ],
    ...,
    [0.00200253],
    [0.00200253],
    [0.00200253]]), 'coupling.2.nu': array([[1.30000000e-05],
    [1.30000000e-05],
    [1.30000000e-05],
    ...,
    [1.29998715e-05],
    [1.29998715e-05],
    [1.29998715e-05]]), 'coupling.2.nutilde': array([[1.30000000e-05],
    [1.30000000e-05],
    [1.30000000e-05],
    ...,
    [1.29971456e-05],
    [1.29971456e-05],
    [1.29971456e-05]]), 'pop.2.q': array([[11.07967784],
    [11.07967784],
    [11.07967785],
    ...,

我试图将其转换为一个熊猫df，其中包含列标题的每个dict键和每个相应的dict值作为其数据序列。谢谢你的帮助。

Answer 1

这就是你要找的吗？

from numpy import array

data = {
    "coupling.2.b": array([[0.00247262], [0.00248247], [0.00249233], [0.01106487], [0.01106487], [0.01106487]]),
    "coupling.2.ca": array([[9.98543405e-09], [9.97095234e-09], [9.95652127e-09], [1.22356842e-08], [1.22356843e-08], [1.22356843e-08]]),
    "coupling.2.gnmda": array([[0.002], [0.002], [0.002], [0.00200253], [0.00200253], [0.00200253]]),
    "coupling.2.nu": array([[1.30000000e-05], [1.30000000e-05], [1.30000000e-05], [1.29998715e-05], [1.29998715e-05], [1.29998715e-05]]),
    "coupling.2.nutilde": array([[1.30000000e-05], [1.30000000e-05], [1.30000000e-05], [1.29971456e-05], [1.29971456e-05], [1.29971456e-05]]),
}

data = {k: v.flatten() for k, v in data.items()}
df = pd.DataFrame(data)

print(df)
   coupling.2.b  coupling.2.ca  coupling.2.gnmda  coupling.2.nu  coupling.2.nutilde
0      0.002473   9.985434e-09          0.002000       0.000013            0.000013
1      0.002482   9.970952e-09          0.002000       0.000013            0.000013
2      0.002492   9.956521e-09          0.002000       0.000013            0.000013
3      0.011065   1.223568e-08          0.002003       0.000013            0.000013
4      0.011065   1.223568e-08          0.002003       0.000013            0.000013
5      0.011065   1.223568e-08          0.002003       0.000013            0.000013

这里的诀窍是通过numpy.ndarray.flatten方法将所有数组都压平。一旦你这样做，它就会无缝地传递到一个数据文件中。