How to Group and Order Pandas Data by Original Appearance – 6 Clever Methods

Introduction

The author presents a Pandas data‑processing challenge: given a data column, create a new column that lists the values in the order they first appear, preserving duplicates. The initial DataFrame and expected output are shown.

import pandas as pd

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)
# Please add your code. The "new" column should contain the grouped‑and‑ordered result.
print(df)

Resulting output is displayed in the following image:

Method 1

A solution contributed by "猫药师Kelly" is shown in the image below.

Method 2

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

The resulting DataFrame is shown in the following screenshot:

Method 3

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 screenshot:

Method 4

import pandas as pd

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

Result screenshot:

Method 5

import pandas as pd

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

Result screenshot:

Method 6

import pandas as pd

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

Result screenshot:

Conclusion

The article showcases six distinct Pandas techniques for generating a column that reflects the original order of elements in another column, illustrating each approach with full code and visual output, and encourages readers to experiment and share alternative solutions.