15 self-contained notebooks tackling real-world data challenges with Pandas/NumPy. Part of the Python Summer Party by Interview Master × DataCamp.
This challenge covers diverse business scenarios: social media analytics, e-commerce, fraud detection, loyalty programs, product launches, and more.
- Data cleaning: Type fixes, missing values, de-duplication, email domains, date normalization
- Filtering & joins: Time windows, multi-condition masks, dimension merges
- Analytics & KPIs: GroupBy aggregations, CTR calculations, segment analysis
- Transformations: Pivot tables, temperature bins, calculated business metrics (revenue per ride, average order value)
- Quality checks: Outlier detection (IQR), data validation
- Day 1: Group size and engagement on filtered subsets.
- Day 2: CTR by category vs overall; identify standouts.
- Day 3: Per-visit spend, first–last deltas, spend tiers.
- Day 4: Remove duplicates; best results-per-page by average time.
- Day 5: Clean data; monthly totals; simple growth projection.
- Day 6: Temperature bins; pivoted sales totals; IQR outlier check.
- Day 7: Quarter slice; missing sales flags; top collaborations.
- Day 8: Payment-method counts/means; what-if sales lift.
- Day 9: Fix dates; percentiles; high-activity users.
- Day 10: Clean ratings; category mean/median/std.
- Day 11: Fraud detection basics: email domain extraction, impute for averages, weekday of high risk.
- Day 12: Identify return customers; monthly return counts.
- Day 13: Deduplicate; flavor averages; per-rating deviation.
- Day 14: Loyalty vs non-loyalty: counts, AOV, % difference.
- Day 15: Earnings-per-mile; averages for higher-occupancy rides.
- Install and launch: pip install jupyterlab pandas numpy && jupyter lab
- Open any .ipynb file and run all cells
I've created a Pandas cheat sheet with 21 practical snippets based on the first week of this challenge. Download it from here
