Movie Ratings Analysis

Collaborative workshop project: clean a movie dataset, explore genre trends, analyze profitability, and train a predictive ratings model.

• Full workshop specification: see docs/SPEC.md
• Environment/setup: concise instructions live in SETUP.md

Quick Start

uv sync
uv run python scripts/01_clean_data.py
uv run python scripts/02_analyze_genres.py
uv run python scripts/03_analyze_financials.py
uv run python scripts/04_build_model.py

The scripts are designed to run in order; each writes its outputs for the next step.

Pipeline Overview

• 00_refresh_raw.py – optional refresh of the TMDB subset (data/movies_raw.csv).
• 01_clean_data.py – feature engineering -> results/movies_clean.csv.
• 02_analyze_genres.py – decade/genre area chart -> outputs/genres_by_decade.png.
• 03_analyze_financials.py – ROI & profitability summary -> outputs/roi_by_budget_category.png.
• 04_build_model.py – scikit-learn regression with cross-val + holdout metrics.

Key Artifacts

• Clean dataset: results/movies_clean.csv
• Plots: outputs/genres_by_decade.png, outputs/roi_by_budget_category.png
• Model metrics: printed by scripts/04_build_model.py

Repository Layout

movie-analysis/
├── README.md
├── SETUP.md
├── docs/
│   └── SPEC.md
├── data/
│   ├── README.md
│   └── movies_raw.csv
├── scripts/
├── outputs/
├── results/
└── tests/

Tips

• Commit after each script so teammates can re-run and review.
• Document notable findings (ROI shifts, genre insights, feature importances) in your PR or the shared report.
• Need more context? The spec in docs/SPEC.md covers roles, timeline, and stretch goals.