[Python] How to subtract data between columns that have same subfix

I have a sample dataframe as below that has same subfix as 001, 002, 003.

import pandas as pd
import numpy as np

branch_names = [f"Branch_{i}" for i in range(1, 11)]
date_1 = '20241231'
date_2 = '20250214'
date_3 = '20250220'

data = {
'Branch': branch_names,
date_1 + '_001': np.random.randint(60, 90, 10),
date_1 + '_002': np.random.randint(60, 90, 10),
date_1 + '_003': np.random.randint(60, 90, 10),
date_2 + '_001': np.random.randint(60, 90, 10),
date_2 + '_002': np.random.randint(60, 90, 10),
date_2 + '_003': np.random.randint(60, 90, 10),
date_3 + '_001': np.random.randint(60, 90, 10),
date_3 + '_002': np.random.randint(60, 90, 10),
date_3 + '_003': np.random.randint(60, 90, 10)
}

# Chuyển thành DataFrame
df = pd.DataFrame(data)




Now I want to subtract data between columns that have same subfix as below principle:

df['diff_1_001'] = df[date_3 + '_001'] - df[date_2 + '_001']
df['diff_2_001'] = df[date_3 + '_001'] - df[date_1 + '_001']
df['diff_1_002'] = df[date_3 + '_002'] - df[date_2 + '_002']
df['diff_2_002'] = df[date_3 + '_002'] - df[date_1 + '_002']
df['diff_1_003'] = df[date_3 + '_003'] - df[date_2 + '_003']
df['diff_2_003'] = df[date_3 + '_003'] - df[date_1 + '_003']
df


As you see that we have same 001, 002, 003 but prefix is different. So I want don't want to hard code the 001, 002, 003 but automatically subtract it as mentioned above.

Continue reading...