MLflow — tracking & registry¶
DVC versions the files: which exact CSV, which exact model binary, which exact .dvc pointer went into a run. MLflow versions the story and manages the lifecycle: every training run's hyperparameters, metrics, and output files (Tracking), a shared server that a whole team can query (backend + artifact stores), and a Model Registry that turns a pile of run artifacts into a governed catalog of named, versioned models with a promotion workflow (Staging → Production) that serving reads by reference. …
🔒 La suite est en accès freemium — lecture complète 100 % gratuite
Tu lis ici l'aperçu libre. Le reste du chapitre (code, schémas, maths, exercices) fait partie du livre complet : crée un compte gratuit (30 secondes, aucun paiement) pour tout lire.