Convert a dictionary to a pandas dataframe

Question

I'm trying to convert a dictionary that only has 1 record to a pandas dataframe. I've used the following code from other solutions:

d =  {'id': 'CS2_056', 'cost': 2, 'name': 'Tap'}

pd.DataFrame(d.items(), columns=['id', 'cost','name'])

But I get the following error:

PandasError: DataFrame constructor not properly called!

Possible duplicate of [Python Dictionary to Pandas Dataframe](http://stackoverflow.com/questions/34589332/python-dictionary-to-pandas-dataframe) — McGrady, Feb 26 '17 at 09:08

Serenity · Answer 1 · 2017-08-24T11:16:51.503

55

You dict has only one record use list:

import pandas as pd
d =  {'id': 'CS2_056', 'cost': 2, 'name': 'Tap'}
df = pd.DataFrame([d], columns=d.keys())
print df

Output:

   id  cost name
0  CS2_056     2  Tap

edited Aug 24 '17 at 11:16

answered Feb 26 '17 at 09:14

Serenity

35,289
20
120
115

This works. I took my original solution from another question but maybe the reason for it not working was because it only had 1 record? – user3302483 Feb 26 '17 at 09:30
2

have been doinf d instead of [d] for almost 30 minutes, thanks for the solution. – im_bhatman Oct 18 '19 at 11:49

score 8 · Answer 2 · answered Nov 01 '17 at 11:34

8

Might be you are using python3. in python3 we have list there

pd.DataFrame(list(d.items()), columns=['id', 'cost','name'])

answered Nov 01 '17 at 11:34

Imran Ahmad Ghazali

605
1
10
16

I had a dictionary of lists, and I wanted every list to be a cell value under one column, not a series of columns. This gave me an idea how to solve that. – Rafs Aug 21 '23 at 09:36

score 6 · Answer 3 · answered Apr 07 '18 at 22:02

An alternative way to create a dataframe with a single row from a dictionary is by creating an empty dataframe first and then appending to it:

import pandas as pd 
d =  {'id': 'CS2_056', 'cost': 2, 'name': 'Tap'}
df = pd.DataFrame().append(d, ignore_index=True)
print(df)

   cost       id name
0   2.0  CS2_056  Tap

Note that this method is significantly slower than @Serenity 's solution so definitely do not choose this method if you are concerned about performance. But having options is always nice.

it seems this method also arranges the columns in alpha ascending order, fwiw. — tb., Jul 04 '23 at 23:04

score 5 · Answer 4 · answered Mar 17 '21 at 15:32

5

Easy command, just make sure you enclose your dictionary in square brackets.

d =  {'id': 'CS2_056', 'cost': 2, 'name': 'Tap'}

df = pd.DataFrame([d])
df

answered Mar 17 '21 at 15:32

DougR

3,196
1
28
29

score 1 · Answer 5 · edited May 23 '17 at 12:17

While this question does have a duplicate (Python Dictionary to Pandas Dataframe), I believe there's a simplier answer than those provided there.

Convert the values to lists:

d = {'id': ['CS2_056'], 'cost': [2], 'name': ['Tap']}

then simply:

df = pd.DataFrame(d)
print(df)
#     cost       id name
#  0     2  CS2_056  Tap

Keep in mind that if columns order matter you'd still need to explicitly provide columns to DataFrame:

df = pd.DataFrame(d, columns=['id', 'cost', 'name'])
print(df)
#          id  cost name
#  0  CS2_056     2  Tap

Vishakha Patil · Answer 6 · 2021-03-01T06:28:25.330

-1

Passing an index parameter to the DataFrame constructor works in python 3.

import pandas as pd

d =  {'id': 'CS2_056', 'cost': 2, 'name': 'Tap'}
df = pd.DataFrame(d, index=[0])
print(df)

# Output:
#           id  cost name
#   0  CS2_056     2  Tap

edited Mar 01 '21 at 06:28

answered Jan 20 '21 at 11:07

Vishakha Patil

59
1
3

Convert a dictionary to a pandas dataframe

6 Answers6

Linked