Introduction

Introduction

Training a good model is like building a good car. It still has to be delivered to customers, serviced regularly, and repaired when something goes wrong. A model that sits in a notebook helps nobody.

Once a model is in use, the world keeps changing — new customers, new prices, new behaviour — and its accuracy slowly drops unless someone looks after it. This final topic of the project walks through the three jobs that start after training:

1. deploying the model so it can be used;
2. monitoring it while it's running;
3. maintaining it with retraining and safe updates.