CSV → Excel Report Builder (with Summary + Charts)

This script converts labeled transaction data from CSV into a multi-sheet Excel report with summaries and charts.

It builds on PyFi’s new AI for Finance demo, extending it into a full reporting pipeline that exports a polished Excel workbook with embedded visualizations and summary statistics.

Click here to learn more about PyFi's Python <> OpenAI integration, and register for the next live demo

---

Background

PyFi’s AI for Finance demo executes logic technically impossible in Excel, injects intelligence into the workflow via the OpenAI API, and automates a multi-hour financial analysis, in just a few seconds.

This project takes the output of that demo and produces a stakeholder-ready Excel report:

• Raw labeled data
• Human-readable summaries
• Embedded chart images
• Native Excel charts built directly from the data

This makes it suitable for finance, ops, or audit-style reporting workflows.

---

What it generates

Output workbook (example: labeled_data.xlsx) includes:

1) labeled_data

Full import of labeled_data.csv produced by the PyFi demo pipeline.

2) Summary

If these files exist, they are added as formatted sections:

• summary.csv
  • expected column: Summary
• summary_stats.csv
  • expected columns: Metric, Value

3) Charts

Embeds pre-rendered PNG charts:

• spending_by_spender.png
• spending_by_category.png
• spending_by_vendor.png

Charts are scaled and positioned side-by-side with titles.

4) Native Charts

Creates Excel-native charts from aggregated data:

• Pie — Spending by Spender
• Pie — Spending by Category
• Bar (top 15) — Spending by Vendor

All charts are generated directly inside Excel using openpyxl.

---

Input expectations

Required

• labeled_data.csv (from PyFi demo output)

Expected columns:

• amount
• spender
• category
• vendor

Optional

• summary.csv
• summary_stats.csv
• PNG chart files listed above

---

Installation

pip install pandas openpyxl