Reshaping 4×1 DataFrame into 2×2 DataFrame

Question:

I have a DataFrame as below:

import pandas as pd

df_testcase = pd.DataFrame(
    ['16SCSE102014', '15/03/2019', '16SCSE101350', '15/03/2019']
)

              0
0  16SCSE102014
1    15/03/2019
2  16SCSE101350
3    15/03/2019

I need to convert it into 2×2 DataFrame like this:

              0           1
0  16SCSE102014  15/03/2019
1  16SCSE101350  15/03/2019

Can someone please help?

Asked By: Komal

||
Source

Answers:

I’m not sure whether there is a trivial way to do that using pandas dataframe. But you can do it using numpy as such:

import pandas as pd
import numpy as np
df = pd.DataFrame({'the_only_column':[1,'a',2,'b',3,'c']})

np_array = np.array(df['the_only_column'])
np_array = np.reshape(np_array, (np_array.shape[0]//2, 2))

df = pd.DataFrame(np_array, columns=['new_column_name', 'other_column_name'])
print(df)

And the output is:

  new_column_name other_column_name
0               1                 a
1               2                 b
2               3                 c
Answered By: Kevin Winata

You can alter the shape of your data using reshape after converting the DataFrame df2 to a numpy ndarray with .values then convert the resulting array back into a DataFrame:

# If res is your (4,1) DataFrame, you can just do
res = df2.values.reshape(2,2)
df = pd.DataFrame(res)
Answered By: Akaisteph7