如何将 dataframe 线拆分为多个数据帧？

Question

I have a dataframe:

    0   1   2   3   4   5  6
0   A   B   C   D   E   F  G
1   H   I   J   K   L   M  N
2   O   P   Q   R   S   T  U
3   V   W   X   Y   Z

I want to split every line into multiples line in the random condition(it can be any condition):

For example,

df['2'],df['4],df['6]
df['0'],df['3']
df['1'],df['5']

In this case, these three rows should be repeated for every row in the input data frame.

Expected output:

C   E   G
A   D
B   F
J   L   N
H   K
I   M
Q   S   U
O   R
P   T
X   Z
V   Y
W
   #should repeat for other rows too

Headers are not required or I can ignore them while converting to csv.

Answer 1

You can specify columns names in list, then in list comprehension filter it and convert columns to default range columns names by DataFrame.set_axis , join by concat , sorting by DataFrame.sort_index , replace missing values and create default index:

vals = [['2','4','6'], ['0','3'], ['1','5']]

L = [df.loc[:, x].set_axis(range(len(x)), axis=1) for x in vals]
df = pd.concat(L).sort_index(kind='mergesort').fillna('').reset_index(drop=True)
print (df)
    0  1  2
0   C  E  G
1   A  D   
2   B  F   
3   J  L  N
4   H  K   
5   I  M   
6   Q  S  U
7   O  R   
8   P  T   
9   X  Z   
10  V  Y   
11  W