Feature/data reconciliation

[RubyRyn]

Inventory Reconciliation

Compares two inventory snapshots taken a week apart, identifies what changed, and outputs a structured reconciliation report.

Deliverables

• ExploratoryDataAnalysis.ipynb — Exploratory data analysis notebook documenting all data quality findings
• reconcile.py — Reconciliation script with data cleaning and merge logic
• tests/test_reconcile.py — pytest coverage for all core functions
• output/ — Reconciliation report in CSV and JSON
• NOTES.md — Write-up on approach, findings, and decisions

Data Quality Issues Found

• Column names differ between snapshots (e.g., qty vs quantity, warehouse vs location)
• SKU formatting inconsistencies (missing hyphens, mixed case)
• Duplicate SKU in snapshot 2 with conflicting data
• Negative quantity value
• Mixed date formats
• Product name mismatch for same SKU

Reconciliation Categories

Each item is classified as: removed, added, increased, decreased, or unchanged.

How to run

python reconcile.py
python -m pytest tests/test_reconciliation.py -v

Notes

Full details on assumptions and decisions are in NOTES.md.