How to Transform a Pandas DataFrame into a Dictionary

This is one useful tip especialy when you want to return your dataframe as a JSON file when you are working with flask. There are 3 ways to do it.

import pandas as pd

df=pd.DataFrame(
{
'Gender':np.random.choice(["m","f"],10,p=[0.6,0.4]),
'Age':np.random.choice(["[<30]","[30-65]", "[65+]"],10,p=[0.3,0.6,0.1]),
    }
)
df

  Gender      Age
0      m  [30-65]
1      m  [30-65]
2      f  [30-65]
3      m  [30-65]
4      m    [<30]
5      m  [30-65]
6      m    [<30]
7      f  [30-65]
8      f  [30-65]
9      f    [<30]

Dict

This is the most known method.

df.to_dict()

{'Gender': {0: 'm',
  1: 'm',
  2: 'f',
  3: 'm',
  4: 'm',
  5: 'm',
  6: 'm',
  7: 'f',
  8: 'f',
  9: 'f'},
 'Age': {0: '[30-65]',
  1: '[30-65]',
  2: '[30-65]',
  3: '[30-65]',
  4: '[<30]',
  5: '[30-65]',
  6: '[<30]',
  7: '[30-65]',
  8: '[30-65]',
  9: '[<30]'}}

The problem here is that it keeps the index in the dictionary and in most cases this is not what we want. We will show you 2 more methods that are not using the index.

Records

df.to_dict('r')

[{'Gender': 'm', 'Age': '[30-65]'},
 {'Gender': 'm', 'Age': '[30-65]'},
 {'Gender': 'f', 'Age': '[30-65]'},
 {'Gender': 'm', 'Age': '[30-65]'},
 {'Gender': 'm', 'Age': '[<30]'},
 {'Gender': 'm', 'Age': '[30-65]'},
 {'Gender': 'm', 'Age': '[<30]'},
 {'Gender': 'f', 'Age': '[30-65]'},
 {'Gender': 'f', 'Age': '[30-65]'},
 {'Gender': 'f', 'Age': '[<30]'}]

List

df.to_dict('l')

{'Gender': ['m', 'm', 'f', 'm', 'm', 'm', 'm', 'f', 'f', 'f'],
 'Age': ['[30-65]',
  '[30-65]',
  '[30-65]',
  '[30-65]',
  '[<30]',
  '[30-65]',
  '[<30]',
  '[30-65]',
  '[30-65]',
  '[<30]']}