/ml-ops-zoomcamp

Primary LanguageJupyter Notebook

ML Ops Zoomcamp

Week 1 - Intro

Notes

  • What is MLOps
  • MLOps maturity model
  • Running example: NY Taxi trips dataset
  • Why do we need MLOps
  • Course overview

Week 2 - Experiment Tracking

Notes

  • Experiment tracking intro
  • Getting started with MLflow
  • Experiment tracking with MLflow
    • Run hyperparameter tuning using Hyperopt, and then logging all the run results in MLflow
  • Saving and loading models with MLflow
  • Promoting best model to Model registry
    • Filter out some of the best performing models on validation sets, run them through test set, thereby promoting the best model to model registry

Week 3 - Orchestration and ML Pipelines

Notes

  • Workflow orchestration
  • Prefect 2.0
    • Concept of task and flow
    • Work-queues and agents
    • Previewing work-queues and view scheduled runs
    • IntervalSchedule vs CronSchedule
  • Created a mini project that downloaded, ingested, trained and validated on datasets automatically based on a scheduled interval time period.

Week 4 - Deploying Models with Flask and Docker

  • Creating a new virtual environment
  • Create a script for prediction
  • Put script into Flask app
  • Dockerize the app

credits: DataTalksCub

author: chekwei