CODEWITHSUNDEEP
pythonpython-3.xpandasdataframedata-manipulation
Zakyl
Zakyl

Reputation: 63

Sort dataframe based on custom dictionary

The dataframe that I have is:

df = pd.DataFrame(data={'Question':['Q2','Q2','Q1','Q1','Q1','Q3','Q3','Q3'],
                    'Answer':['Yes','No','$1 to $49','$100 to $200','$50 to $100','More than 5000','Less than 5000','Don't know']})

I would like to sort the dataframe by the column Question and Answer. I have created a custom dictionary to be used when sorting by Answer, so that categorical values can be sorted accordingly.

answer_sort_order = {'$1 to $49': 0, '$50 to $100': 1, '$50 to $99': 2, '$100 to $200': 3,'More than 5000': 4, 'Less than 5000': 5, 'Don't Know': 6}

How can I use this to get the dataframe like below?

enter image description here

I can also specify that to only use the answer_sort_order dictionary for records in which Question is Q1 and Q3

Upvotes: 0

Views: 359

Answers (1)

mozway
mozway

Reputation: 260725

You can use the key parameter of sort_values:

out = df.sort_values('Answer', key=pd.Series(answer_sort_order).reindex)

or:

out = df.sort_values('Answer', key=lambda x: x.map(answer_sort_order))

output:

  Question          Answer
2       Q1       $1 to $49
4       Q1     $50 to $100
3       Q1    $100 to $200
5       Q3  More than 5000
6       Q3  Less than 5000
0       Q2             Yes
1       Q2              No
7       Q3      Don't know

Upvotes: 1

Related Questions