Python Pandas 列按计数排序

Question

我对 python 和 pandas 还是个新手。我想弄清楚如何根据每列的行数对数据进行排序。不包括 NaN 的行数是：

np.count_nonzero(~np.isnan(df.iloc[:, j])，其中 j 是列号

import pandas as pd
x = np.nan
data = {"A": [22133.97, 22151.06, 22836.96, 22035.42, 23306.39, 23881.29, 23206.37], "B": [95.924, 107.1005, 123.5775, 107.8946, x, x, x], "C": [74.169, 74.075, 77.564, 76.338, 79.356, 81.666, x], "D": [36.205, 35.435, 36.542, 34.424, 37.457, x, x], "E": [68.048, 65.554, 68.093, 68.37, 74.233, 77.095, 75.156]}
dates = pd.date_range('1/1/2000', periods = 7)
df = pd.DataFrame(data, index = dates)

数据集现在看起来像这样：

                   A         B       C       D       E
2000-01-01  22133.97   95.9240  74.169  36.205  68.048
2000-01-02  22151.06  107.1005  74.075  35.435  65.554
2000-01-03  22836.96  123.5775  77.564  36.542  68.093
2000-01-04  22035.42  107.8946  76.338  34.424  68.370
2000-01-05  23306.39       NaN  79.356  37.457  74.233
2000-01-06  23881.29       NaN  81.666     NaN  77.095
2000-01-07  23206.37       NaN     NaN     NaN  75.156

是否有任何内置的 pandas 函数来对列进行排序，如下所示（即 A 列有 7 行，所以它仍然存在，然后搜索下一个最列E 并将其向前移动并将其放在 A 列旁边，依此类推）。

                   A       E       C       D         B
2000-01-01  22133.97  68.048  74.169  36.205   95.9240
2000-01-02  22151.06  65.554  74.075  35.435  107.1005
2000-01-03  22836.96  68.093  77.564  36.542  123.5775
2000-01-04  22035.42  68.370  76.338  34.424  107.8946
2000-01-05  23306.39  74.233  79.356  37.457       NaN
2000-01-06  23881.29  77.095  81.666     NaN       NaN
2000-01-07  23206.37  75.156     NaN     NaN       NaN

非常感谢任何帮助。谢谢！

Answer 1

我认为这段代码可以工作。

df=df[df.isna().sum().sort_values().keys()]

输出:

>>>df[df.isna().sum().sort_values().keys()]
                   A       E       C       D         B
2000-01-01  22133.97  68.048  74.169  36.205   95.9240
2000-01-02  22151.06  65.554  74.075  35.435  107.1005
2000-01-03  22836.96  68.093  77.564  36.542  123.5775
2000-01-04  22035.42  68.370  76.338  34.424  107.8946
2000-01-05  23306.39  74.233  79.356  37.457       NaN
2000-01-06  23881.29  77.095  81.666     NaN       NaN
2000-01-07  23206.37  75.156     NaN     NaN       NaN

Python Pandas 列按计数排序

Python Pandas Column Sort by Count

python

columnsorting