Feature Normalization and Percentage of Group Total with transform()

Question

Given `products` (product_name, category, sales), add a new column `category_sales_share` representing each product's sales divided by the total sales of its category.

Accepted Answer

Optimal Solution:
import pandas as pd

# transform('sum') returns a Series aligned with the original DataFrame index
category_totals = df.groupby('category')['sales'].transform('sum')

df['category_sales_share'] = (df['sales'] / category_totals).round(2)

result_df = df

Technical Explanation:
`transform('sum')` evaluates the group sum but retains the original DataFrame row count and index alignment, avoiding clumsy merge operations.

product_name	category	sales
iPhone	Phones	8000
Pixel	Phones	2000
MacBook	Laptops	15000

product_name	category	sales	category_sales_share
iPhone	Phones	8000	0.8
Pixel	Phones	2000	0.2
MacBook	Laptops	15000	1