Start: 18 August 2021

104 - Machine Learning Pipelines Automation, reproducible experiments and MLOps (ENG)

Boost Machine Learning experiments and engineering with DVC, MLflow, Airflow and other Open Source tool
Write your awesome label here.

What's included?

  • 6 learning modules
  • 1 end-to-end course project 
  • 100 % online
  • lessons recording
  • code review

Hands-On Project

You will work on batch scoring project during the course. 
Learn DVC, MLflow and Airflow by integrating them for the course project purposes.

From Jupyter Notebooks to automated MLOps

Start from a prototype in Jupyter Notebook and develop fully automated pipelines and MLOps. Step by step. 

Should you enroll? 

Novice in DS / ML

You will learn essential engineering skills you need to grow in Data Science & Machine Learning.
You will learn what is required in real projects, boost your technical skills and increase your value for the company

Data Scientists

Improve engineering skills for automating ML experiments, documentation and reports generation. Step forward to developing of production solutions and speed up ML experiments

Machine Learning Engineers

The course will be interesting if you are approaching to integrate DVC, MLflow, Airflow, but do not have enough experience with them. Or you need to quickly figure out how to integrated and make them work together

Course Program

1. Organize your project & code

Overview approaches and technologies that helps to organize work on Machine Learning (ML) projects, code and teamwork. Set up a repository, review requirements for team collaboration with Git, toolkit for tracking tasks, hypotheses and changes in an ML project

2. Manage environment dependencies with Python virtual environments and Docker

Let's deal with Docker and docker-compose. Set up development environment for an Machine Learning project

3. Version data and automate pipelines with DVC

Get started with versioning data, artifacts and models and pipelines automation with Data Version Control (DVC). We automate the pipeline for training models and assessing their quality. After that you may run ML experiments with only one command!

4. ML experiments management and metrics tracking with DVC and MLflow

Let's add MLflow to our project! Now we have an nice UI for tracking metrics and parameters of experiments, comparing experiments, visualizing the results of GridSearch, etc. DVC and MLflow are used together to manage experiments and model lifecycle

5. Automate pipelines with Airflow

Let's get started with Airflow! What you can use it for? How to create pipelines? How to integrate it with DVC and MLflow? Airflow is often used for production run models for batch scoring on a schedule. This is a good solution for running forecast generation in batch mode

6. Setup CI/CD and MLOps for your ML solution with DVC, Gitlab, Arflow & MLFlow

Setting up an automatic CI / CD process using for our Machine Learning solutions (MLOps). We apply DVC, Gitlab, Arflow & MLFlow tools. Also, let's add monitoring of our system using Grafana and Prometheus

Course Program

Meet the instructor

Mikhail Rozhkov

Data Scientist & Machine Learning Engineer

Co-Creator of the Machine Learning REPA project. Has over 6 years hands-on experience in Machine Learning & Data Science, leads projects and helps teams to implement good tools and engineering practices
Patrick Jones - Course author

Course Packs

* We will refund your money if, within 14 days after the start of the course, you realize that it is not suitable for you!

Starting

 you pay by yourself on site
USD 500
  • Online lessons & course materials access
  • Code Examples
  • Standard Course Project (table data, scoring)
  • Course Discussion Chat

Professional

 you pay by yourself on site
USD 750
  • Online lessons & course materials access
  • Code Examples
  • Standard Course Project (table data, scoring)
  • Course Discussion Chat
  • Guided Course Project
  • Code Review
  • Weekly Office Hours Discussion Sessions 

Corporate

your company pays by invoice
USD 1250
  • Online lessons & course materials access
  • Code Examples
  • Standard Course Project (table data, scoring)
  • Course Discussion Chat
  • Guided Course Project
  • Code Review
  • Weekly Office Hours Discussion Sessions 
  • Your work project / use case discussion (on demand)

Course outcomes

Work confidently with Git

Organize your project code and be confident in collaboration with Git

Automate pipelines

Automate pipelines for experiments, data preparation and model evaluation with DVC

Setup CI/CD

Set up continuous integration and delivery processes with GitLab

Manage models and metrics

Manage experiments, metrics and model lifecycle with DVC and MLflow

Organize a team work

Apply GitFlow и CodeReview in team collaboration on Machine Learning projects

Setup MLOps

Use Airflow for MLOps tasks, configure monitoring of the work of models with Grafana

Deploy to Production

Develop a production solutions and run scoring batch jobs by schedule

Documenting ML experiments

Organize tools for automated reporting and docs generation 

Do you need a group workshop?

If you are interesting in group workshops on the course materials for your company, please contact us!
Created with