Machine Learning (ML) · Advanced level

Advanced level · Module 4 of 6 · 24 guided hours

Deploying Machine Learning Models

Package models reproducibly and serve predictions through APIs or batch jobs that others can run and trust.

Learning objectives

  • Save models and environments reproducibly
  • Serve predictions through a validated web API
  • Choose between batch and real-time deployment and package services with containers

Topics

  • saving pipelines with joblib
  • environment and dependency pinning
  • FastAPI services
  • input validation with Pydantic
  • batch vs real-time prediction
  • Docker basics

Lessons

  1. Lesson 1 Saving Models Reproducibly

    Save trained pipelines and record everything needed to load and trust them later.

    joblibdependency pinningrandom seedsmodel metadata

  2. Lesson 2 Serving Predictions with an API

    Wrap a model in a web API with validated inputs, clear errors and a health check.

    FastAPIPydantic schema422 validation errorshealth endpoint

  3. Lesson 3 Batch and Real-Time Deployment

    Choose between scheduled batch predictions and on-demand APIs based on how predictions are used.

    batch scoringreal-time inferencelatencyscheduling

  4. Lesson 4 Packaging with Containers

    Package the service and its dependencies in a container so it runs the same way everywhere.

    containerDockerfileimage sizeconfiguration through environment variables

Practical activity

Serve the price forecasting model as a FastAPI service with input validation and a health check, and also as a monthly batch job that writes a report.

Expected outcomes

  • Deployable, reproducible ML services
  • Sound deployment choices

Assessment

Practical: the service must run on the instructor's machine from the README and reject invalid input with clear errors.

Taught live by an instructor — online or in person. Free course lessons are for self-study and do not lead to a certificate — see registration options.