Pivot and melt

pivot / pivot_table widen data (long → wide); melt tidies data (wide → long). Essential for report-style tables and tidy-data pipelines.

pivot_table

import pandas as pd
df = pd.DataFrame({'date': ['Jan','Jan','Feb'],
                   'product': ['A','B','A'],
                   'sales': [10, 20, 15]})
wide = df.pivot_table(index='date', columns='product',
                      values='sales', aggfunc='sum', fill_value=0)
print(wide)

melt

long = wide.reset_index().melt(id_vars='date',
                             var_name='product', value_name='sales')
print(long)

When to use

• pivot_table — cross-tab summaries with aggregation (handles duplicates)
• pivot — strict reshape when index/column pairs are unique
• melt — stack columns into variable/value pairs for plotting or modeling

Important interview questions and answers

1. Q: pivot vs pivot_table?
   A: pivot_table aggregates duplicates; pivot errors on duplicate index/column keys.
2. Q: Tidy data?
   A: One row per observation—melt helps convert wide spreadsheets to tidy form.

Self-check

1. Create a pivot_table of sales by date and product.
2. Melt a wide table back to long format.

Pitfall: Use pivot_table when duplicate index/column pairs exist—strict pivot errors.