33 lines (27 loc) · 1.6 KB

Tennis Data & Model Project

What this is

A small project that collects tennis match data and includes a lightweight prediction model UI/build.
Meant for exploration, quick experiments, and sharing a reproducible dataset + model bundle.

What's inside

A dataset folder with recent Roland Garros match data and engineered CSVs for modelling.
A small frontend/build in pred_model containing the production build and source.
Scripts and examples to run experiments locally.

Quick start

Explore the data: open the dataset CSVs to see match features and engineered columns.
Inspect the frontend: the pred_model folder contains a React app build under pred_model/build and the source in pred_model/src.
Run locally (optional):
- If you want to run the React app from source, install dependencies in pred_model and start the dev server.
- For quick checks, open the prebuilt pred_model/build/index.html in a browser.

How to use these files

Datasets: use the CSVs for training, analysis, or feature experiments.
Model/frontend: the build is ready to serve; if you want to modify the UI, edit the source in the pred_model/src folder and re-build.

Good practices

Keep a copy of raw CSVs and a separate copy for engineered features.
Track experiments (parameters, train/validation splits) in a small log or notebook.
If you modify the frontend, bump the version and note major interface changes.

Need help?

I can:
- Generate a short requirements.txt or package.json updates.
- Create a simple script to load a CSV and show sample rows. Tell me which next step you'd like.