How to Group and Order Data by Appearance in Pandas – 6 Practical Methods

Introduction

In a recent Python community challenge, participants were asked to use Pandas to process a DataFrame so that the elements in the data column are grouped and ordered according to their first occurrence, with the result shown in a new column.

The original DataFrame is displayed below:

Implementation

Method 1

Kelly provided a solution with code and the resulting output.

Method 2

Yuliang offered another answer. The core code is:

df['newnew'] = sum([[k]*v for k, v in Counter(df['data']).items()], [])

Execution produced the following result:

Method 3

Another shared snippet uses itertools.chain:

import pandas as pd
from collections import Counter
from itertools import chain

df = pd.DataFrame({
    'data': ['A1','D3','B2','C4','A1','A2','B2','B3','C3','C4','D5','D3'],
    'new': ['A1','A1','D3','D3','B2','B2','C4','C4','A2','B3','C3','D5']
})
print(df)

df['newnew'] = [*chain(*([k]*v for k, v in Counter(df['data']).items()))]
print(df)

Result:

Method 4

Moon God presented a method similar to the previous ones:

df['new2'] = df['data'].unique().repeat(df['data'].value_counts(sort=False))

Result:

Method 5

Another approach uses categorical ordering:

df['new3'] = df['data'].astype('category').cat.reorder_categories(df['data'].unique()).sort_values().values

Result:

Method 6

The final method sorts the list by the original index order:

df['new4'] = sorted(df['data'].tolist(), key=df['data'].tolist().index)

Result:

Conclusion

The article showcases six distinct Pandas solutions for grouping and ordering a column based on the order of appearance, encouraging readers to explore additional methods and share their own approaches.