Creating dataframe from list of arrays (varying shapes)

Question

I am trying to convert a list of arrays of varying shapes to a dataframe.

import numpy as np
import pandas as pd

data = [np.array([[1, 2], [1, 3], [1, 1]]),
        np.array([[1, 2, 3], [3, 1, 2], [3, 2, 1]])]
names = ['A', 'B']

df = pd.DataFrame(data=data, columns=names)
df

However, this gives the error-

ValueError: Shape of passed values is (2, 1), indices imply (2, 2)

I then tried-

df = pd.DataFrame(np.array([None, *data], dtype=object)[1:]).T
df

                          0                                  1
0  [[1, 2], [1, 3], [1, 1]]  [[1, 2, 3], [3, 1, 2], [3, 2, 1]]

Which is not my desired output.

I want each inner list in as separate rows, like the following:

        A          B
0  [1, 2]  [1, 2, 3]
1  [1, 3]  [3, 1, 2]
2  [1, 1]  [3, 2, 1]

Not sure how to proceed.

Answer 1

Try:

pd.DataFrame(dict((k,list(v)) for k,v in zip(names, data)))

Output:

        A          B
0  [1, 2]  [1, 2, 3]
1  [1, 3]  [3, 1, 2]
2  [1, 1]  [3, 2, 1]

Answer 2

this is what worked for me, istead of sending the data as nasted lists i sended a dictionary which define its values for each column name, this way pandas didnt converted it to 3 columns:

data = [array([[1, 2],
        [1, 3],
        [1, 1]]),
 array([[1, 2, 3],
        [3, 1, 2],
        [3, 2, 1]])]
names = ['A', 'B']
pd.DataFrame({name:l.tolist() for name,l in zip(names,data)})
Out[5]: 
        A          B
0  [1, 2]  [1, 2, 3]
1  [1, 3]  [3, 1, 2]
2  [1, 1]  [3, 2, 1]

the wrong way

pd.DataFrame(data)
>>>
                                  0
0           [[1, 2], [1, 3], [1, 1]]
1  [[1, 2, 3], [3, 1, 2], [3, 2, 1]]

# or

pd.DataFrame([l.tolist() for l  in data])
>>>
           0          1          2
0     [1, 2]     [1, 3]     [1, 1]
1  [1, 2, 3]  [3, 1, 2]  [3, 2, 1] 

Answer 3

Let us try concat , it will do it one by one sub-data

out = pd.concat([pd.Series(list(x)) for x in data], keys=names, axis=1)
        A          B
0  [1, 2]  [1, 2, 3]
1  [1, 3]  [3, 1, 2]
2  [1, 1]  [3, 2, 1]