What is the efficient way of splitting a pandas DataFrame column in 2

Question

I have a pandas DataFrame that looks like this:

               x1y1             x2y2
0    [694.0, 427.0]  [1178.0, 601.0]
1    [621.0, 415.0]   [736.0, 456.0]
2    [551.0, 404.0]   [669.0, 461.0]
3    [514.0, 421.0]   [569.0, 463.0]
4    [181.0, 406.0]   [320.0, 462.0]
5    [738.0, 415.0]   [873.0, 474.0]
6   [1158.0, 446.0]  [1209.0, 513.0]
7    [613.0, 176.0]   [692.0, 272.0]
8      [2.0, 295.0]    [50.0, 368.0]
9    [817.0, 305.0]   [870.0, 373.0]
10  [1130.0, 410.0]  [1174.0, 500.0]
11  [1155.0, 420.0]  [1199.0, 497.0]
12   [990.0, 417.0]  [1053.0, 524.0]
13   [952.0, 409.0]  [1003.0, 515.0]
14   [905.0, 412.0]   [944.0, 503.0]
15    [34.0, 432.0]    [84.0, 485.0]
16    [1091.0, 1.0]   [1172.0, 78.0]
17    [859.0, 49.0]   [975.0, 146.0]
18    [710.0, 76.0]   [827.0, 145.0]
19     [68.0, 62.0]   [181.0, 115.0]
20  [1076.0, 252.0]  [1142.0, 297.0]
21  [1058.0, 298.0]  [1103.0, 372.0]
22   [642.0, 336.0]   [675.0, 366.0]
23   [777.0, 382.0]   [800.0, 408.0]
24   [264.0, 241.0]   [331.0, 292.0]

I want to split it into a DataFrame with x1, y1, x2, y2 as columns efficiently without for loops or iterating over the rows, is there some way to do so?

Answers there do not work so far: `pd.DataFrame(frame_before['x1y1'].tolist(), columns=['x1, y1'])` gives ValueError: 1 columns passed, passed data had 2 columns — , Apr 20 '20 at 10:05
But is this an efficient way? I mean is creating a new DataFrame for each column efficient? — , Apr 20 '20 at 10:09

score 2 · Answer 1 · answered Apr 20 '20 at 10:16

Idea is use numpy.hstack to 2d array and pass to DataFrame constructor:

df3 = pd.DataFrame(np.hstack((df['x1y1'].tolist(), 
                              df['x2y2'].tolist())), columns=['x1', 'y1', 'x2', 'y2'])
print (df3)
        x1     y1      x2     y2
0    694.0  427.0  1178.0  601.0
1    621.0  415.0   736.0  456.0
2    551.0  404.0   669.0  461.0
3    514.0  421.0   569.0  463.0
4    181.0  406.0   320.0  462.0
5    738.0  415.0   873.0  474.0
6   1158.0  446.0  1209.0  513.0
7    613.0  176.0   692.0  272.0
8      2.0  295.0    50.0  368.0
9    817.0  305.0   870.0  373.0
10  1130.0  410.0  1174.0  500.0
11  1155.0  420.0  1199.0  497.0
12   990.0  417.0  1053.0  524.0
13   952.0  409.0  1003.0  515.0
14   905.0  412.0   944.0  503.0
15    34.0  432.0    84.0  485.0
16  1091.0    1.0  1172.0   78.0
17   859.0   49.0   975.0  146.0
18   710.0   76.0   827.0  145.0
19    68.0   62.0   181.0  115.0
20  1076.0  252.0  1142.0  297.0
21  1058.0  298.0  1103.0  372.0
22   642.0  336.0   675.0  366.0
23   777.0  382.0   800.0  408.0
24   264.0  241.0   331.0  292.0

What is the efficient way of splitting a pandas DataFrame column in 2

1 Answers1