Overview

Orientation to CoopRecSys, its recommendation model families, production data stack, native acceleration, and documentation map.

How CoopRecSys validates Python code, Cython extensions, Ary2Tower backend selection, wheel contents, and release artifacts.

A release workflow for building native wheels, validating Ary2Tower, tagging releases, and publishing CoopRecSys artifacts.

A practical first-run path covering data preparation, model selection, training, exact top-N inference, and evaluation.

How CoopRecSys uses qrates to turn implicit transactions into pseudo-ratings, weighted scores, and feature-group retrieval signals.

How CoopRecSys separates data access, feature engineering, configuration, model families, native kernels, fallback logic, scoring, and reporting.

How the noisemaker package supports controlled train/test splits, feature exchange, and robustness experiments without changing the model implementation.

How CoopRecSys structures group-aware ranking, feature-rich recommendation, tuning, inference, and experiment tracking with LightGBM.

Install CoopRecSys from a wheel or source checkout, verify the native stack, and build Ary2Tower extensions locally when required.

Overview of CoopRecSys reporting, visualization, dashboard surfaces, configuration inspection, and model diagnostics.

Detailed guidance for prepare, db, features, encoders, DuckDB, and reusable feature pipelines in CoopRecSys.

Reference for the current ./src/cooprecsys package structure and the responsibilities of its major directories.

Development workflow, model/API compatibility rules, native build expectations, documentation standards, and regression testing for CoopRecSys.

File-by-file reference for the shared CoopRecSys configuration directory and the model-local configuration pattern.

Detailed guide to configuration.ini, typed configuration classes, fallback settings, LightGBM settings, and logging in CoopRecSys.

How CoopRecSys uses the assets package for dashboard utilities, chart data, miniature statistics, and packaged frontend resources.

Technical guide to the AryColBring model family, supported losses, native kernels, evaluation, and wheel usage.

Production guide to Ary2Tower training, Cython/OpenMP inference, exact top-N behavior, and the modern residual fallback.

How to import CoopRecSys, which layers are public integration boundaries, and how to keep training/inference code portable.